CN111930737B - Multidimensional correlation analysis method for equipment combat test data - Google Patents

Multidimensional correlation analysis method for equipment combat test data Download PDF

Info

Publication number
CN111930737B
CN111930737B CN202011089603.5A CN202011089603A CN111930737B CN 111930737 B CN111930737 B CN 111930737B CN 202011089603 A CN202011089603 A CN 202011089603A CN 111930737 B CN111930737 B CN 111930737B
Authority
CN
China
Prior art keywords
data
equipment
combat
test
test 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.)
Active
Application number
CN202011089603.5A
Other languages
Chinese (zh)
Other versions
CN111930737A (en
Inventor
刘会英
王凯
蒲玮
张仁友
孙俊峰
谢奇
彭文成
唐正华
陈财森
陈志敏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Academy of Armored Forces of PLA
Original Assignee
Academy of Armored Forces of PLA
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Academy of Armored Forces of PLA filed Critical Academy of Armored Forces of PLA
Priority to CN202011089603.5A priority Critical patent/CN111930737B/en
Publication of CN111930737A publication Critical patent/CN111930737A/en
Application granted granted Critical
Publication of CN111930737B publication Critical patent/CN111930737B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/21—Design, administration or maintenance of databases
    • G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2465—Query processing support for facilitating data mining operations in structured databases
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/24—Querying
    • G06F16/245—Query processing
    • G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
    • G06F16/2474—Sequence data queries, e.g. querying versioned data
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2216/00—Indexing scheme relating to additional aspects of information retrieval not explicitly covered by G06F16/00 and subgroups
    • G06F2216/03—Data mining

Abstract

The invention provides a multidimensional correlation analysis method for equipment combat test data, which comprises the following steps of constructing a combat test data content system: cleaning and classifying combat test data; selecting and preparing data; preprocessing static data; preprocessing dynamic time sequence data; forming a set of transaction items; performing association analysis according to a strong association rule; useful knowledge information is extracted from the strong association rule according to the relevant prior knowledge of the combat test, and decision support information is provided for the compilation of a combat test scheme, the equipment compilation and the equipment quality identification. The multidimensional association analysis method for the equipment combat test data can analyze the association among various combat test data, and can provide guidance for design of combat test schemes, test action reply, innovation of combat test methods and the like.

Description

Multidimensional correlation analysis method for equipment combat test data
Technical Field
The invention relates to the technical field of equipment combat tests, in particular to a multidimensional correlation analysis method for equipment combat test data.
Background
The combat test refers to equipment test activities for examining and evaluating equipment combat effectiveness, guarantee effectiveness, army applicability, combat task satisfaction, quality stability and the like according to equipment combat mission tasks and combat processes under approximate actual combat conditions. At present, with the continuous development of various combat test activities of the whole army, massive equipment combat test data are accumulated. And the equipment combat test data are collected for analysis, so that reliable data support can be provided for equipment combat test conclusion.
However, more emphasis is placed on statistical analysis based on the collected data in the battle test activities to draw conclusions such as "whether the equipment battle efficiency can meet the battle demand, how applicable the army of the equipment is, how stable the quality of the equipment is", and the like. However, the relevance analysis of various types of massive combat test data is rarely involved, and the potential value hidden behind the massive combat test data cannot be deeply excavated. At present, the understanding of the correlation analysis of combat test data at home and abroad is limited, a set of mature and reliable theory and method is not formed, and related concepts and theoretical methods are still in the research and discussion stage.
The journal of China Equipment engineering 4 in 2020 discloses an equipment combat test data analysis preprocessing method of Xiaje, Tianoyu and Wang. The data conversion and classification solve the problems of data classification and storage, the accuracy of data is ensured by abnormal value detection, and the data processing is more visual and convenient by non-dimensionalized parameter processing. On the basis, experimental data are counted by using a small sample theory, and verification analysis is carried out by combining examples. The method is only suitable for processing discrete numerical data and cannot process a large amount of time sequence data existing in a combat test. In addition, the method adopts the classical small sample statistical theory T distribution and T test, aims to make up the defect of small sample amount under the small sample test condition, and is not suitable for the application scene of mass data.
Disclosure of Invention
In order to solve the technical problems, the equipment combat test data multidimensional association analysis method provided by the invention can be used for analyzing the association among various combat test data and providing guidance for combat test scheme design, test action reply, combat test method innovation and the like.
The invention aims to provide a multidimensional association analysis method for equipment combat test data, which comprises the following steps of constructing a combat test data content system:
step 1: cleaning and classifying combat test data;
step 2: selecting and preparing data;
and step 3: preprocessing static data;
and 4, step 4: preprocessing dynamic time sequence data;
and 5: forming a set of transaction items;
step 6: performing association analysis according to a strong association rule;
and 7: useful knowledge information is extracted from the strong association rule according to the relevant prior knowledge of the combat test, and decision support information is provided for the compilation of a combat test scheme, the equipment compilation and the equipment quality identification.
Preferably, the constructing of the combat test data content system comprises dividing the combat test data into equipment basic data, test environment data, combat configuration data, combat action data, test command data, equipment operator data and equipment running state data.
In any of the above schemes, preferably, the step 1 includes performing missing value processing and abnormal value processing on the combat test data, and performing corresponding classification and carding on the combat test data according to the previously constructed combat test data content system.
In any of the above schemes, preferably, the step 2 includes selecting the interested data set by screening the classified and grouped combat test data according to the data mining purpose.
In any of the above aspects, preferably, the static data includes equipment basis data, test environment data, combat orchestration data, and equipment operator data in the data set.
In any of the above aspects, preferably, the static data preprocessing method includes the following sub-steps:
step 31: carrying out data normalization, and unifying data items which have the same meaning but are expressed differently into the same description language;
step 32: extracting keywords of a descriptive language;
step 33: and removing useless items and repeated items, and deleting repeated items or items without actual distinguishing effect on the records.
In any of the above aspects, preferably, the dynamic time series data includes test command data, operational action data, and equipment operational status data in a data set.
In any of the above schemes, preferably, the dynamic time series data preprocessing method includes the following sub-steps:
step 41: dividing data time slices, carrying out time slicing on continuous time sequence data according to corresponding business requirements, and preserving the characteristics of time sequence operation data to the maximum extent while discretizing the data;
step 42: discretizing the time series data characteristics, wherein the time series data characteristics comprise the change trend of the time series data besides the self value of the time series data;
step 43: and performing emergency transaction setting, and setting a transaction item for the equipment with an emergency.
In any of the above schemes, preferably, the step 5 further includes sorting the data in each record in the battle trial data set, and the set of each data item forms a complete transaction set I = { I = { I = }1,i2,i3,…ik… } in which ikFor a transaction, a transaction item X is formed for each record, and X belongs toIEach X includes a plurality of ikAll records are taken as a collection of transaction itemsT={X 1 ,X 2 ,X 3 ,…X k ,…}。
In any of the above schemes, preferably, the strong association rule is generated for the transaction item set T by:
step 61: scanning the transaction item set T, determining the support degree of each transaction, and obtaining all frequent-1 item sets L1;
step 62: find frequent "2 item sets" set C2 according to L1;
and step 63: cutting off items which do not meet the support degree threshold value to obtain L2;
step 64: find frequent "3 item sets" set C3 according to L2;
step 65: pruning is carried out according to the property and the support degree threshold value to obtain L3;
and step 66: looping steps 61 through 65 until an empty set is obtained, i.e., until a larger frequent set L cannot be found;
step 67: and calculating the non-empty subsets of the maximum frequent set L, and calculating the confidence coefficient pairwise to obtain the strong association rule larger than the threshold value of the confidence coefficient.
The invention provides a multidimensional correlation analysis method for equipment combat test data, which can effectively make up the defects that the correlation analysis of various kinds of massive combat test data in the current equipment combat test is basically blank and the potential value hidden behind the massive combat test data cannot be deeply excavated, and can effectively improve the combat test capability of an equipment system and the post analysis level of the combat test data.
Drawings
Fig. 1 is a flowchart of a preferred embodiment of a multidimensional correlation analysis method for equipment combat trial data according to the present invention.
Fig. 2 is a flowchart of an embodiment of a static data preprocessing method of the multidimensional correlation analysis method for equipment combat test data according to the present invention.
Fig. 3 is a flowchart of an embodiment of a dynamic time series data preprocessing method for a multidimensional correlation analysis method of equipment combat test data according to the present invention.
Fig. 4 is a flowchart of an embodiment of a method for generating strong association rules of equipment test data according to the multidimensional association analysis method of the equipment combat test data.
Fig. 5 is a flowchart of another preferred embodiment of the multidimensional correlation analysis method for equipment combat trial data according to the present invention.
Fig. 6 is a content architecture diagram of an embodiment of equipment combat test data according to the multi-dimensional correlation analysis method of the equipment combat test data of the present invention.
Fig. 7 is a flowchart of another embodiment of a method for generating strong association rules of equipment test data according to the equipment combat test data multidimensional association analysis method of the present invention.
Detailed Description
The invention is further illustrated with reference to the figures and the specific examples.
Example one
As shown in fig. 1, step 100 is executed to construct a combat test data content system, and divide the combat test data into equipment basic data, test environment data, combat configuration data, combat action data, test command data, equipment operator data, and equipment operating state data.
And step 110, cleaning and classifying the combat test data, processing missing values and abnormal values of the combat test data, and correspondingly classifying and carding the combat test data according to the combat test data content system constructed in the prior art.
And step 120, selecting and preparing data, screening the classified and grouped combat test data according to the data mining purpose, and selecting an interested data set.
Step 130 is performed, static data preprocessing, wherein the static data includes equipment basis data, test environment data, combat configuration data, and equipment operator data in the data set. As shown in fig. 2, in this step, step 131 is executed to perform data normalization, and unify data items having the same meaning but different expressions into the same description language. Step 132 is executed to extract keywords of the descriptive language. Step 133 is executed to remove the useless items and the duplicate items, and delete the duplicate items or the items that have no actual distinguishing effect on the records.
Step 140 is executed to preprocess dynamic time series data, wherein the dynamic time series data comprises test command data, operational action data and equipment operational state data in a data set. As shown in fig. 3, in this step, step 141 is executed, data time slice division is performed, time slice division is performed on continuous time sequence data according to corresponding business requirements, and time sequence operation data characteristics are retained to the maximum extent while data are discretized. Step 142 is executed to perform discretization processing on the time series data characteristics, wherein the characteristics of the time series data include the variation trend of the time series data besides the self value of the time series data. Step 143 is executed to perform emergency transaction setting, and set a transaction item for the equipment emergency.
Executing step 150, forming a transaction item set, sorting the data in each record in the combat trial data set, wherein the set of each data item forms a transaction complete set I = { I = }1,i2,i3,…ik… } in which ikFor a transaction, a transaction item X is formed for each record, and X belongs toIEach X includes a plurality of ikAll records are taken as a collection of transaction itemsT={X 1 ,X 2 ,X 3 ,…X k ,…}。
Step 160 is executed to perform association analysis according to the strong association rule. As shown in fig. 4, in this step, step 161 is executed to scan the transaction item set T, determine the support of each transaction, and obtain all the frequent-1 item sets L1. Step 162 is performed to find the frequent "2 item sets" set C2 from L1. Step 163 is performed to prune the entries that do not meet the support threshold, resulting in L2. Step 164 is performed, finding frequent "3 item sets" set C3 from L2. Step 165 is executed, pruning is performed according to the property and the support degree threshold, and L3 is obtained.
Step 166 is performed to determine whether the resulting set is an empty set, i.e., until a larger frequent set L cannot be found. If the resulting set is not an empty set, step 161 is re-executed. If the obtained set is an empty set, step 167 is executed to calculate the non-empty subset of the maximum frequent set L, and calculate the confidence level pairwise, so as to obtain the strong association rule larger than the threshold value of the confidence level.
And step 170, extracting useful knowledge information from the strong association rule according to the relevant prior knowledge of the combat test, and providing decision support information for the compilation of the combat test scheme, the equipment compilation and the equipment quality identification.
Example two
A method for analyzing the association between the data of different combat tests features that the data of different combat tests are analyzed to provide guidance for the design of scheme, action and reply of test, and the innovation of combat test.
As shown in fig. 5, a method for analyzing the data association of a combat test includes the following steps:
the method comprises the following steps: as shown in fig. 6, a battle trial data content system was constructed. The combat test data is divided into seven types including equipment basic data, test environment data, combat configuration data, combat action data, test command data, equipment operator data and equipment running state data.
Step two: and cleaning and classifying the combat test data. And carrying out missing value processing and abnormal value processing on the combat test data, and correspondingly classifying and carding the combat test data according to the combat test data content system constructed in the prior art.
Step three: and selecting and preparing data. Aiming at the specific analysis problem, screening is carried out on classified and grouped combat test data according to the data mining purpose, and an interested data set is selected.
Step four: and (5) preprocessing static data. Static data such as equipment basic data, test environment data, combat configuration data and equipment operator data in the data set are processed as follows:
(1) and (6) normalizing the data. Unifying the same meaning but different data items into the same description language.
(2) Keywords of a descriptive language are extracted.
(3) Useless items and repeated items are removed. Duplicate items or items that do not actually distinguish between records are deleted.
Step five: and preprocessing dynamic time sequence data. The following processing is carried out on dynamic time sequence data such as test command data, combat action data, equipment combat operation state data and the like in the data set:
(1) and dividing the data time slice. And carrying out time scribing on the continuous time sequence data according to corresponding business requirements, and preserving the time sequence operation data characteristics to the maximum extent while discretizing the data. For example, at time t1, the travel Speed of certain equipment satisfies 0 ≦ Speed ≦ 30KmWhen setting a transaction to IS0；30≦Speed﹤45KmIs set as transaction IS1And so on. For another example, at time t1, the task performed by a certain type of equipment is motorization march, and the set transaction is ITask0(ii) a The running task is an impact and the transaction is set as ITask1。
(2) And discretizing the time sequence data characteristics. The characteristics of the time series data include the variation trend of the time series data in addition to the value of the time series data itself. For example, at time t2, the travel Speed of certain equipment is less than or equal to 30 ≦ Speed ≦ 45KmInternal, set the transaction to IS1If the value rises from time t1, the timing characteristic transaction may be set to Iup。
(3) Setting an emergency transaction. The transaction item is mainly set for the equipment with an emergency. Such as tkAt the moment a certain type of equipment fails, a transaction can be set to ItroubleAnd can be at tk-1Setting a transaction to I at a timetrouble-beforeAt tk+1Setting a transaction to I at a timetrouble-after。
Step six: a set of transaction items is formed. On the basis of the fourth step and the fifth step, the data in each record in the combat trial data set are sorted, and the set of each data item forms a complete transaction set I = { I =1,i2,i3,…,ik… } in which ikIs a transaction. Forming a transaction item X for each record, wherein X belongs to the groupI. Each X contains a plurality of ik. Taking all records as a set of transaction items T = { X1,X2,X3,…,Xk,…}。
Step seven: and (5) correlation analysis. The strong association rule is generated for the set of transaction items T by the following method. The method for generating the strong association rule of the equipment test data is shown in fig. 7.
(1) Scanning a transaction item set T for the first time, determining the support degree of each transaction, and obtaining all frequent-1 item sets L1;
(2) find frequent "2 item sets" set C2 according to L1;
(3) cutting off items which do not meet the support degree threshold value to obtain L2;
(4) find frequent "3 item sets" set C3 according to L2;
(5) pruning is carried out according to the property and the support degree threshold value to obtain L3;
(6) the above process is cycled until an empty set is obtained, i.e. until a larger frequent set L cannot be found;
(7) and calculating the non-empty subsets of the maximum frequent set L, and calculating the confidence coefficient pairwise to obtain the strong association rule larger than the threshold value of the confidence coefficient.
Step eight: useful knowledge information is extracted from the strong association rule according to the relevant prior knowledge of the combat test, and decision support information is provided for the compilation of a combat test scheme, the equipment compilation and the equipment quality identification.
The invention has the following advantages:
1. a combat test data content system is established, a framework is provided for classification and carding of combat test data, and the efficiency of data association analysis can be improved.
2. A method for preprocessing dynamic time sequence data of a combat test is provided, and a transaction item set can be generated efficiently.
3. The provided method for analyzing the association of the combat test data can deeply mine the association rule knowledge contained in massive combat test data and provide decision support for design of combat test schemes and quality evaluation of equipment.
EXAMPLE III
A method for analyzing operational test data association comprises the following steps:
1. and cleaning and classifying the combat test data.
And carrying out missing value processing and abnormal value processing on the combat test data, and correspondingly classifying and carding the combat test data according to the combat test data content system constructed in the prior art.
2. A problem is defined.
As shown in table 1, the proposed operational test correlation analysis method is mainly used herein to find correlations between the operational state of equipment and equipment basic data, test environment data, operational configuration data, operational action data, test command data, equipment operator data, and the like, with respect to a certain type of equipment.
TABLE 1 definition of problem set
Figure DEST_PATH_IMAGE001
3. And selecting and preparing data.
Aiming at the defined problems, 65345 pieces of equipment application state data of certain type of equipment in the combat test are selected, and the associated data items comprise equipment rated maximum driving speed, rated maximum climbing gradient, rated engine power, rated maximum range, equipment production time, equipment service time, equipment health state, weather data, road data, terrain data, equipment marshalling data, task execution type, task execution target, task execution progress, task execution personnel, test command, equipment operators and the like.
4. And (4) preprocessing data.
And (3) preprocessing static data according to the method in the step 4: carrying out normalization processing on the equipment running state data: the engine power is reduced, the engine power is insufficient, the engine oil is insufficiently combusted, and the like are unified into the engine working abnormity; extracting key words from equipment fault description sentences, converting descriptive language into description characteristics and analyzing the description characteristics, wherein the extracted key words comprise 'track breakage', 'engine can not ignite', 'machine gun can not excite', 'observation and aiming system fault', 'power box fault' and the like. Setting up transactions such as weather IweatherTopography IterrainPerson type Itype。
And (3) preprocessing the dynamic time series data according to the method in the step 5: time scribing was performed in units of 30 min. Setting transactions, e.g. speed of travel IsHair-care deviceOperating state of motive machine IengineOil temperature IoilWater temperature IwaterTo perform a combat action Iaction(ii) a Setting time-series characteristic transactions, e.g. speed-down Isoeed_downIncrease in running speed Ispeed_upDecreasing range Ishootdistance_downEtc.; setting incident transactions, e.g. tkAt the moment when a certain type of equipment has an engine fault, a transaction can be set as Itrouble_engineAnd can be at tk-1Setting a transaction to I at a timetrouble_engine-beforeAt tk+1Setting a transaction to I at a timetrouble_engine-afterAnd the like. After data preprocessing, a transaction set in the form of table 2 can be obtained.
Table 2 transaction attribute encoding
Figure 938774DEST_PATH_IMAGE002
5. Forming a collection of transaction items
After the equipment running state data is preprocessed, the equipment running state of each time slice is 1 event, 1 transaction item is generated by combining the previous transaction set, and finally, a transaction item set T is generated.
6. Association analysis
And performing association analysis on the transaction item set T, and setting the minimum support degree to be 10% and the minimum confidence degree to be 50%. The correlation results are sorted, obvious useless items are removed, and the obtained correlation results are shown in table 3.
TABLE 3 correlation analysis results
Figure DEST_PATH_IMAGE003
7. Extracting relevant knowledge of combat test
And extracting useful knowledge information from a strong association rule obtained from association analysis according to the prior knowledge of the equipment combat test, and providing decision support information for design of the combat test and quality identification of the equipment.
For a better understanding of the present invention, the foregoing detailed description has been given in conjunction with specific embodiments thereof, but not with the intention of limiting the invention thereto. Any simple modifications of the above embodiments according to the technical essence of the present invention still fall within the scope of the technical solution of the present invention. In the present specification, each embodiment is described with emphasis on differences from other embodiments, and the same or similar parts between the respective embodiments may be referred to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and for the relevant points, reference may be made to the partial description of the method embodiment.

Claims (8)

1. A multidimensional correlation analysis method for equipment combat test data comprises the steps of constructing a combat test data content system, and is characterized by further comprising the following steps:
step 1: cleaning and classifying combat test data;
step 2: selecting and preparing data;
and step 3: preprocessing static data;
and 4, step 4: preprocessing dynamic time sequence data; the dynamic time series data preprocessing method comprises the following substeps:
step 41: dividing data time slices, carrying out time slicing on continuous time sequence data according to corresponding business requirements, and preserving the characteristics of time sequence operation data to the maximum extent while discretizing the data;
step 42: discretizing the time series data characteristics, wherein the time series data characteristics comprise the change trend of the time series data besides the self value of the time series data;
step 43: setting an emergency affair, and setting an affair item aiming at the occurrence of an emergency condition of equipment;
and 5: forming a set of transaction items;
step 6: performing association analysis according to a strong association rule; generating the strong association rule for a set of transaction items T by:
step 61: scanning the transaction item set T, determining the support degree of each transaction, and obtaining all frequent-1 item sets L1;
step 62: find frequent "2 item sets" set C2 according to L1;
and step 63: cutting off items which do not meet the support degree threshold value to obtain L2;
step 64: find frequent "3 item sets" set C3 according to L2;
step 65: pruning is carried out according to the property and the support degree threshold value to obtain L3;
and step 66: looping steps 61 through 65 until an empty set is obtained, i.e., until a larger frequent set L cannot be found;
step 67: calculating the non-empty subsets of the maximum frequent set L, and calculating confidence coefficients pairwise to obtain a strong association rule larger than a confidence coefficient threshold;
and 7: useful knowledge information is extracted from the strong association rule according to the relevant prior knowledge of the combat test, and decision support information is provided for the compilation of a combat test scheme, the equipment compilation and the equipment quality identification.
2. The method for multidimensional correlation analysis of equipment operational test data according to claim 1, wherein the constructing of the operational test data content system comprises dividing the operational test data into equipment basis data, test environment data, operational configuration data, operational action data, test command data, equipment operator data and equipment operational status data.
3. The multidimensional correlation analysis method for the equipment combat test data as claimed in claim 2, wherein the step 1 comprises the steps of carrying out missing value processing and abnormal value processing on the combat test data, and carrying out corresponding classification and combing on the combat test data according to a previously constructed combat test data content system.
4. The multidimensional correlation analysis method for the equipment combat test data as recited in claim 3, wherein the step 2 comprises the step of selecting the interested data set by screening the classified and grouped combat test data according to the data mining purpose.
5. The method of multidimensional correlation analysis of equipment operational test data as recited in claim 4, wherein the static data includes equipment base data, test environment data, operational orchestration data, and equipment operator data in a data set.
6. The method for multidimensional correlation analysis of rig operational test data as recited in claim 5, wherein the static data preprocessing method comprises the substeps of:
step 31: carrying out data normalization, and unifying data items which have the same meaning but are expressed differently into the same description language;
step 32: extracting keywords of a descriptive language;
step 33: and removing useless items and repeated items, and deleting repeated items or items without actual distinguishing effect on the records.
7. The method for multidimensional correlation analysis of equipment combat trial data as recited in claim 6 wherein the dynamic timing data comprises trial command data, combat action data and equipment combat application status data in a data set.
8. The method for multidimensional correlation analysis of equipment combat trial data as recited in claim 7, wherein said step 5 further comprises collating data in each record in the combat trial data set, each set of data items forming a complete set of transactions I ═ { I ═ I { [ where1,i2,i3,…ik… } in which ikFor a certain transaction, a transaction item X is formed for each record, and X belongs to I, wherein each X comprises a plurality of IkAll records are taken as a set of transaction items T ═ X1,X2,X3,…Xk,…}。
CN202011089603.5A 2020-10-13 2020-10-13 Multidimensional correlation analysis method for equipment combat test data Active CN111930737B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202011089603.5A CN111930737B (en) 2020-10-13 2020-10-13 Multidimensional correlation analysis method for equipment combat test data

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202011089603.5A CN111930737B (en) 2020-10-13 2020-10-13 Multidimensional correlation analysis method for equipment combat test data

Publications (2)

Publication Number Publication Date
CN111930737A CN111930737A (en) 2020-11-13
CN111930737B true CN111930737B (en) 2021-03-23

Family

ID=73334465

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202011089603.5A Active CN111930737B (en) 2020-10-13 2020-10-13 Multidimensional correlation analysis method for equipment combat test data

Country Status (1)

Country Link
CN (1) CN111930737B (en)

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20120137367A1 (en) * 2009-11-06 2012-05-31 Cataphora, Inc. Continuous anomaly detection based on behavior modeling and heterogeneous information analysis
CN106503097A (en) * 2016-10-14 2017-03-15 国政通科技股份有限公司 A kind of method and system for improving the quality of data

Also Published As

Publication number Publication date
CN111930737A (en) 2020-11-13

Similar Documents

Publication Publication Date Title
CN110223168B (en) Label propagation anti-fraud detection method and system based on enterprise relationship map
US20060179051A1 (en) Methods and apparatus for steering the analyses of collections of documents
CN111340063B (en) Data anomaly detection method for coal mill
Gürbüz et al. Data mining and preprocessing application on component reports of an airline company in Turkey
CN111882446A (en) Abnormal account detection method based on graph convolution network
CN111365015A (en) Shield tunneling parameter feature extraction and attitude deviation prediction method based on XGboost
Cuzzocrea et al. An information-theoretic approach for setting the optimal number of decision trees in random forests
CN110908957A (en) Network security log audit analysis method in power industry
CN115794803B (en) Engineering audit problem monitoring method and system based on big data AI technology
CN107194617A (en) A kind of app software engineers soft skill categorizing system and method
CN115238154A (en) Search engine optimization system
CN115544519A (en) Method for carrying out security association analysis on threat information of metering automation system
CN103425748B (en) A kind of document resources advise the method for digging and device of word
CN109471934B (en) Financial risk clue mining method based on Internet
CN111930737B (en) Multidimensional correlation analysis method for equipment combat test data
CN113722230B (en) Integrated evaluation method and device for vulnerability mining capability of fuzzy test tool
Homocianu et al. An Analysis of Scientific Publications on'Decision Support Systems' and'Business Intelligence'Regarding Related Concepts Using Natural Language Processing Tools
CN113515560A (en) Vehicle fault analysis method and device, electronic equipment and storage medium
CN106777191A (en) A kind of search modes generation method and device based on search engine
CN111967911A (en) Derivative monitoring and analyzing method and system
Kristiana et al. Association Rule Implementation Using Algorithm Apriori To Analize Fishing Pattern In Indonesia
Liang et al. Shearer cutting pattern recognition based on multi-scale fuzzy entropy and support vector machine
CN113379093A (en) Energy consumption analysis and optimization method for oil gas gathering and transportation system
Tao et al. Research on fault diagnosis method of axial flow induced draft fan of power plant based on machine learning
CN117829435B (en) Urban data quality management method and system based on big 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
GR01 Patent grant
GR01 Patent grant