CN113282645A - Satellite time sequence parameter analysis method, system, terminal and storage medium - Google Patents
Satellite time sequence parameter analysis method, system, terminal and storage medium Download PDFInfo
- Publication number
- CN113282645A CN113282645A CN202110834911.4A CN202110834911A CN113282645A CN 113282645 A CN113282645 A CN 113282645A CN 202110834911 A CN202110834911 A CN 202110834911A CN 113282645 A CN113282645 A CN 113282645A
- Authority
- CN
- China
- Prior art keywords
- sequence
- satellite
- frequent
- frequent pattern
- length
- 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
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
- G06F16/24—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
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Software Systems (AREA)
- Computational Linguistics (AREA)
- Position Fixing By Use Of Radio Waves (AREA)
Abstract
The application relates to a satellite time sequence parameter analysis method, a satellite time sequence parameter analysis system, a terminal and a storage medium. The method comprises the following steps: acquiring a satellite parameter data set; knowledge mining is carried out on the satellite parameter data set by adopting an incidence relation analysis algorithm to obtain a frequent pattern among sequences of all satellite timing sequence parameters in the satellite parameter data set; and generating an association rule of the satellite parameter data set according to the inter-sequence frequent pattern, and predicting the change trend of the satellite parameters in the set time in the future according to the association rule. According to the method and the device, knowledge mining is carried out on the satellite parameter sequences to obtain the inter-sequence frequent patterns of all the satellite parameter sequences, association rules of the satellite parameter sequences are mined according to the inter-sequence frequent patterns, multiple real-time satellite parameters can be analyzed, the defect that permission cannot be given to the multi-time sequence data of the in-orbit satellite in the prior art is overcome, and reliable decision support is provided for safe and stable operation of the satellite.
Description
Technical Field
The present application relates to the field of satellite data analysis technologies, and in particular, to a method, a system, a terminal, and a storage medium for analyzing satellite timing parameters.
Background
With the continuous development of the commercial aerospace industry, the problem of safe operation of on-orbit satellites becomes a research hotspot. The parameters transmitted to the ground by the satellite through the sensor are time sequence data with the characteristics of multiple types, large quantity and the like. By analyzing the incidence relation among the telemetering parameters, the change trend of the telemetering parameters in a short time can be predicted, the satellite parameters causing the abnormal change can be traced through the abnormal change trend of the telemetering parameters, and reliable decision support is provided for safe and stable operation of the satellite. However, the existing association analysis is directed at transaction data sets, and cannot be applied to the association analysis of the on-orbit satellite multi-time series data.
Disclosure of Invention
The application provides a satellite time sequence parameter analysis method, a satellite time sequence parameter analysis system, a satellite time sequence parameter analysis terminal and a storage medium, and aims to solve the technical problem that an incidence relation analysis method in the prior art cannot be applied to incidence relation analysis of multi-time sequence data of an in-orbit satellite.
In order to solve the above problems, the present application provides the following technical solutions:
a satellite timing parameter analysis method comprises the following steps:
acquiring a satellite parameter data set;
knowledge mining is carried out on the satellite parameter data set by adopting an incidence relation analysis algorithm to obtain a frequent pattern among sequences of all satellite timing sequence parameters in the satellite parameter data set;
and generating an association rule of the satellite parameter data set according to the inter-sequence frequent pattern, and predicting the change trend of the satellite parameters in the set time in the future according to the association rule.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the acquiring a satellite parameter dataset further comprises:
and carrying out time alignment, data reduction and data discretization on the satellite parameter sequence in the satellite parameter data set, and converting the satellite parameter sequence into a set data format.
The technical scheme adopted by the embodiment of the application further comprises the following steps:
the time alignment specifically comprises: deleting all satellite parameter sequences at a certain time point in the satellite parameter data set if one or more satellite parameter sequences are missing at the certain time point;
the data reduction specifically comprises the following steps: a sliding window w is given by using a segment aggregation approximation method, the mean value of all satellite parameter sequences in the current sliding window is calculated from the first satellite parameter sequence in the satellite parameter data set, and the mean value is used as the representative value of the satellite parameter sequences in the current sliding window; then, sliding backwards by a step length with the size of w, calculating the mean value of all satellite parameter sequences in the next sliding window, and taking the mean value as the representative value of the satellite parameter sequences in the next sliding window; repeating the steps until the calculation of all the satellite parameter sequences in the satellite parameter data set is completed;
the data discretization specifically comprises the following steps: capturing the variation trend between two adjacent satellite parameter sequences in the satellite parameter data set by adopting a monotonic characteristic extraction method, dividing the variation trend according to a set threshold value, and representing various variation trends through discretization symbols; the set thresholds comprise a normal _ ths threshold used for dividing the normal variation trend and an abrormal _ ths threshold used for dividing the abnormal variation trend; the normal change trend comprises constant level, normal rising and normal falling; the abnormal change trend comprises rapid rising and rapid falling.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the knowledge mining of the satellite parameter data set by using the incidence relation analysis algorithm to obtain the inter-sequence frequent pattern of all the satellite timing sequence parameters in the satellite parameter data set specifically comprises:
performing in-sequence frequent pattern mining on the satellite parameter data set to generate an in-sequence frequent pattern set of all satellite timing sequence parameters in the satellite parameter data set;
and mining inter-sequence frequent patterns of the satellite parameter data set according to the intra-sequence frequent pattern set to generate an inter-sequence frequent pattern set of all satellite timing sequence parameters in the satellite parameter data set.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the mining of the inter-sequence frequent pattern of the satellite parameter data set according to the intra-sequence frequent pattern set specifically comprises:
given a minimum support count min _ s, a maximum support count max _ s, a maximum length windowsize:
selecting a frequent pattern IFPS _1 with the length of 1 in a sequence, wherein the IFPS _1 is composed of discretization symbols;
connecting every two discretization symbols in the in-sequence frequent pattern IFPS _1 to generate a candidate frequent pattern with the length of 2, scanning a satellite parameter data set, and finding and recording a position list of the candidate frequent pattern in the sequence; wherein, the position list records the starting position and the ending position of the candidate frequent pattern in each sequence;
judging whether the length of the position list of the candidate frequent patterns in the sequence is between the minimum support degree min _ s and the maximum support degree count max _ s, and if the length of the position list is between the minimum support degree min _ s and the maximum support degree count max _ s, adding the candidate frequent patterns in the sequence into a frequent pattern set in the sequence; otherwise, deleting the candidate frequent patterns in the sequence;
selecting two intra-sequence frequent patterns with the length of k-1, judging whether the sequence without the first discretization symbol of the frequent pattern in the first sequence is the same as the sequence without the last discretization symbol of the frequent pattern in the second sequence, and if the sequences are the same, connecting the discretization symbols of the intra-sequence frequent patterns in the two sequences to generate a candidate frequent pattern with the length of k in the sequence; wherein k is more than or equal to 2;
finding out position lists of two in-sequence frequent patterns in the candidate frequent patterns with the length of k, generating a new position list by using the position lists of the two in-sequence frequent patterns, judging whether the length of the new position list is between a minimum support degree count min _ s and a maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of k into a frequent pattern set in the sequence; otherwise, deleting the candidate frequent patterns with the length of k in the sequence;
and judging whether the length of the frequent pattern in the frequent pattern set in the sequence reaches a set maximum length windowsize, and if the length of the frequent pattern in the frequent pattern set in the sequence does not reach the set maximum length windowsize, regenerating a candidate frequent pattern in the sequence with the length of k until the length of the frequent pattern in the frequent pattern set in the sequence reaches the set maximum length windowsize.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the mining of the inter-sequence frequent pattern of the satellite parameter data set according to the intra-sequence frequent pattern set specifically comprises:
giving an intra-sequence frequent pattern IFPS _1 and an inter-sequence frequent pattern maximum length maxBlocks; defining the position of a frequent pattern among sequences as the starting position of the first frequent pattern and the last frequent pattern in the frequent pattern set in the sequences;
connecting the frequent modes among different satellite parameter sequences in the intra-sequence frequent mode IFPS _1 pairwise from the intra-sequence frequent mode IFPS _1 to generate a candidate frequent mode with the length of 2 among the sequences;
generating a position list of the candidate frequent patterns with the length of 2 among the sequences through the position list of the candidate frequent patterns in the sequences, judging whether the length of the position list is between the minimum support degree count min _ s and the maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of 2 among the sequences into a frequent pattern set among the sequences; otherwise, deleting the inter-sequence candidate frequent pattern with the length of 2;
generating a candidate frequent pattern with the length of k among sequences by using two frequent patterns with the length of k-1;
finding out a position list of two frequent patterns in the candidate frequent patterns with the length of k from the position list of the candidate frequent patterns in the sequence, generating a new position list by using the position lists of the two frequent patterns, judging whether the length of the new position list is between the minimum support degree count min _ s and the maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of k into the inter-sequence frequent pattern set; otherwise, deleting the candidate frequent patterns with the length of k among the sequences;
and judging whether the length of the frequent pattern in the inter-sequence frequent pattern set reaches a set maximum length maxBlock, and if not, regenerating the candidate frequent pattern with the length of k among the sequences until the length of the frequent pattern in the inter-sequence frequent pattern set reaches the set maximum length maxBlock.
The technical scheme adopted by the embodiment of the application further comprises the following steps: the association rule for generating the satellite parameter data set according to the inter-sequence frequent pattern specifically includes:
dividing the inter-sequence frequent pattern into a source parameter frequent pattern source and a target parameter frequent pattern target, and separating the divided last inter-sequence frequent pattern to be used as a quasi-target parameter frequent pattern freq;
calculating confidence of the inter-sequence frequent pattern: conf = freq support count/source support count; wherein the support count is a location list length of the inter-sequence frequent pattern;
and judging whether the confidence conf is greater than the set minimum confidence min _ conf, and if so, outputting the inter-sequence frequent pattern into an association rule form of source- > target.
Another technical scheme adopted by the embodiment of the application is as follows: a satellite timing parameter analysis system, comprising:
a data acquisition module: for obtaining a satellite parameter dataset;
a data mining module: the satellite parameter data set is subjected to knowledge mining by adopting an incidence relation analysis algorithm to obtain a sequence-to-sequence frequent pattern of all satellite timing sequence parameters in the satellite parameter data set;
a rule generation module: and the association rule is used for generating the satellite parameter data set according to the inter-sequence frequent pattern, and predicting the change trend of the satellite parameters in the set time in the future according to the association rule.
The embodiment of the application adopts another technical scheme that: a terminal comprising a processor, a memory coupled to the processor, wherein,
the memory stores program instructions for implementing the satellite timing parameter analysis method;
the processor is configured to execute the program instructions stored by the memory to control data flow connection optimization.
The embodiment of the application adopts another technical scheme that: a storage medium having stored thereon program instructions executable by a processor to perform the satellite timing parameter analysis method.
Compared with the prior art, the embodiment of the application has the advantages that: according to the satellite time sequence parameter analysis method, the satellite time sequence parameter analysis system, the satellite time sequence parameter analysis terminal and the storage medium, the satellite parameter sequence is preprocessed and converted into a data format capable of being subjected to association rule analysis, knowledge mining is conducted on the satellite parameter sequence through an association relation analysis algorithm, inter-sequence frequent patterns of all satellite parameter sequences are obtained, association rule mining of the satellite parameter sequence is conducted according to the inter-sequence frequent patterns, and the change trend of the satellite parameters within a certain time in the future is predicted according to the association rules. The invention can analyze the remote measurement parameters of a plurality of real-time satellites, solves the defect that the prior art cannot permit the multi-time sequence data of the on-orbit satellite, and provides reliable decision support for the safe and stable operation of the satellite.
Drawings
FIG. 1 is a flow chart of a method for analyzing satellite timing parameters according to an embodiment of the present application;
FIG. 2 is a graph of data reduction comparison according to an embodiment of the present application, wherein (a) is a graph of unreduced data and (b) is a graph of reduced data;
FIG. 3 is a schematic diagram of monotonic feature extraction according to an embodiment of the present application;
FIG. 4 is a schematic diagram illustrating a connection manner of frequent patterns in a sequence according to an embodiment of the present application;
FIG. 5 is a schematic diagram of association rules generated in an embodiment of the present application;
FIG. 6 is a schematic structural diagram of a data stream connection optimization system according to an embodiment of the present application;
fig. 7 is a schematic structural diagram of a terminal according to an embodiment of the present application;
fig. 8 is a schematic structural diagram of a storage medium according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present application and are not intended to limit the present application.
Please refer to fig. 1, which is a flowchart illustrating a method for analyzing satellite timing parameters according to an embodiment of the present application. The satellite time sequence parameter analysis method comprises the following steps:
s10: acquiring a plurality of satellite parameter sequences to generate a satellite parameter data set;
s20: preprocessing the satellite parameter data set to generate a data format capable of carrying out association rule analysis;
in this step, the data format that can be analyzed by the association rule has the characteristic of discretization, that is, each point represents a variation trend of the satellite parameter sequence. For example, for the satellite parameter sequence S1, there are the following consecutive data points: uudld, the continuous data point represents that the satellite parameter sequence S1 has the trend of rising (up), rising, falling (down), level (level) and falling during the time. Specifically, the preprocessing of the satellite parameter data set specifically includes the following steps:
s21: time alignment is carried out on the satellite parameter sequence in the satellite parameter data set;
wherein, the time alignment specifically comprises: if one or more satellite parameter sequences are missing at a certain time point, all the satellite parameter sequences at the time point in the data set are deleted, and finally the time point without the satellite parameter sequences is reserved.
S22: performing data reduction on a satellite parameter sequence in the satellite parameter data set;
the data reduction can reduce the data volume, make the data smoother and simultaneously reserve the original data information to the maximum extent. In the embodiment of the present application, the data reduction specifically includes: setting a sliding window w by using a Piecewise aggregation approximation (PAA for short), starting from a first satellite parameter sequence, calculating the mean value of all satellite parameter sequences in the current sliding window, and taking the mean value as the representative value of the satellite parameter sequences in the current sliding window; then, sliding backwards by a step length with the size of w, calculating the mean value of all satellite parameter sequences in the next sliding window, and taking the mean value as the representative value of the satellite parameter sequences in the next sliding window; and so on until the calculation of all the satellite parameter sequences in the data set is completed. Specifically, as shown in fig. 2, the graph is a data reduction comparison graph according to an embodiment of the present application, wherein (a) is a schematic diagram of unreduced data, and (b) is a schematic diagram of reduced data.
S23: carrying out data discretization on a satellite parameter sequence in the satellite parameter data set, and converting the satellite parameter sequence into a data format capable of carrying out association rule analysis;
the data discretization is carried out on the satellite parameter sequence by using a monotonic feature extraction method in the embodiment of the application. The monotonic characteristic extraction method is used for capturing the variation trend between two adjacent satellite parameter sequences and dividing the variation trend according to a set threshold value. Specifically, in the embodiment of the present application, the set thresholds include two, which are respectively normal _ ths used for dividing normal variation trends such as constant level, normal rise, and normal fall, and abrormal _ ths used for dividing abnormal variation trends such as rapid rise and rapid fall.
Fig. 3 is a schematic diagram of monotonic feature extraction according to an embodiment of the present application. The discretization symbols in the graph are respectively used for representing various variation trends, wherein: "l" represents a constant level, "u" represents a normal rise, "d" represents a normal fall, "f" represents a rapid rise, and "s" represents a rapid fall.
S30: knowledge mining is carried out on the preprocessed satellite parameter data set by adopting an incidence relation analysis algorithm to obtain inter-sequence frequent patterns of all satellite parameter sequences in the satellite parameter data set;
in this step, compared with a frequent item set in a classical incidence relation analysis algorithm, the inter-sequence frequent pattern of the embodiment of the present application introduces the sequentiality of the satellite parameter sequence, that is, if the contents of the two frequent patterns are the same but the sequences are different, the two frequent patterns belong to two different patterns.
Specifically, the frequent pattern mining process includes intra-sequence frequent pattern mining and inter-sequence frequent pattern mining, and the specific process is as follows:
s31: mining frequent patterns in the sequence;
s31.1: given a minimum support count min _ s, a maximum support count max _ s, a maximum length windowsize:
s31.2: selecting a length-1 intra-sequence frequent pattern IFPS _1 (i.e., a not-yet-concatenated intra-sequence frequent pattern);
in the present invention, IFPS _1 is all discretization symbols: f, u, l, d, s; starting with length 1, the individual symbols, i.e. 'u' 'l''d ', etc., are the frequent patterns in the sequence with length 1. Since there are a total of 5 discretized symbols, there are a total of 5 frequent patterns within the length-1 sequence.
S31.3: connecting every two discretization symbols in the in-sequence frequent pattern IFPS _1 to generate a candidate frequent pattern with the length of 2, scanning a satellite parameter data set, and finding and recording a position list of the candidate frequent pattern in the sequence; wherein, the start position and the end position of each candidate frequent pattern in the sequence are recorded in the position list, for example, the position list of the candidate frequent pattern "ud" in the sequence is [ [22,23], [60,61], …, [3304,3305] ].
S31.4: and judging whether the position list length (namely, how many times the position list appears) of the candidate frequent patterns in the sequence is between the minimum support degree min _ s and the maximum support degree count max _ s, if the position list length is between the minimum support degree min _ s and the maximum support degree count max _ s, judging the candidate frequent patterns in the sequence as frequent patterns, and adding the candidate frequent patterns in the sequence into the frequent pattern set in the sequence. Otherwise, deleting the candidate frequent patterns in the sequence;
s31.5: selecting two intra-sequence frequent patterns with the length of k-1, judging whether the sequence without the first discretization symbol of the frequent pattern in the first sequence is the same as the sequence without the last discretization symbol of the frequent pattern in the second sequence, and if the sequences are the same, connecting the discretization symbols of the intra-sequence frequent patterns in the two sequences to generate a candidate frequent pattern with the length of k in the sequence; the connection of the frequent patterns within the sequence is shown in fig. 4. Wherein k is more than or equal to 2.
S31.6: scanning a satellite parameter data set, respectively finding and recording position lists of two in-sequence frequent modes in the in-sequence candidate frequent modes with the length of k, generating a new position list by using the position lists of the two in-sequence frequent modes, judging whether the length of the new position list is between the minimum support degree min _ s and the maximum support degree count max _ s, and adding the in-sequence candidate frequent mode with the length of k into the in-sequence frequent mode set if the length of the position list is between the minimum support degree min _ s and the maximum support degree count max _ s; otherwise, deleting the candidate frequent pattern in the sequence with the length of k.
S31.7: judging whether the length of the frequent pattern in the frequent pattern set in the sequence reaches the set maximum length windowsize, if not, repeatedly executing S31.5 and S31.6 until the length of the frequent pattern in the frequent pattern set in the sequence reaches the set maximum length windowsize; if the set maximum length windowsize is reached, all intra-sequence frequent patterns of the satellite parameter data set are obtained. In consideration of the sequence of the discretization symbols, the number of the candidate frequent patterns in the sequence in the embodiment of the application is 5 × 5=25, and the position lists of the candidate frequent patterns in the 25 sequences are respectively found and recorded by scanning the satellite parameter data set; after the position lists of the candidate frequent patterns in all the sequences are obtained, the subsequent frequent pattern mining does not need to repeatedly scan the satellite parameter data set, and only needs to carry out the frequent pattern mining according to the position lists which are continuously iterated, so that the mining time is greatly shortened.
S32: mining inter-sequence frequent patterns according to the intra-sequence frequent pattern set;
s32.1: giving an intra-sequence frequent pattern IFPS _1, a minimum support count min _ s, a maximum support count max _ s and an inter-sequence frequent pattern maximum length maxBlocks; defining the position of one inter-sequence frequent pattern as the starting position of the first frequent pattern and the last frequent pattern in the inter-sequence frequent pattern set;
s32.2: connecting the frequent patterns among different satellite parameter sequences in the IFPS _1 pairwise from the IFPS _1 (the connection of the frequent patterns of the same satellite parameter sequence is meaningless), and generating a candidate frequent pattern with the length of 2 among the sequences;
s32.3: generating a position list of the candidate frequent patterns with the length of 2 among the sequences through the position list of the candidate frequent patterns in the sequences, judging whether the length of the position list is between the minimum support degree min _ s and the maximum support degree count max _ s, and if the length of the position list is between the minimum support degree min _ s and the maximum support degree count max _ s, adding the candidate frequent patterns with the length of 2 among the sequences into a frequent pattern set among the sequences; otherwise, deleting the inter-sequence candidate frequent pattern with the length of 2.
The position list generation mode of the inter-sequence candidate frequent pattern specifically comprises the following steps: two time intervals min _ t and max _ t are given, which respectively represent the minimum interval and the maximum interval of the second inter-sequence candidate frequent pattern after the first inter-sequence candidate frequent pattern appears. The second inter-sequence candidate frequent pattern needs to appear within a time range of (min _ t, max _ t) after the occurrence of the first inter-sequence candidate frequent pattern.
Traversing the position list of the candidate frequent pattern among the first sequences, circularly judging whether the initial position + min _ t of each item is less than the initial position of the corresponding item in the position list of the candidate frequent pattern among the second sequences, if so, continuously judging whether the initial position + max _ t of each item in the candidate frequent pattern among the first sequences is more than or equal to the position of the corresponding item of the candidate frequent pattern among the second sequences, if so, taking the initial position of each item in the candidate frequent pattern among the first sequences as the initial position of the item of the connected frequent pattern, and taking the initial position of the item in the candidate frequent pattern among the second sequences as the end position of the item of the connected frequent pattern; when the circulation is finished, all position lists of the candidate frequent patterns among the connected sequences can be found.
S32.4: generating a candidate frequent pattern with the length of k between the sequences by using two frequent patterns with the length of k-1, wherein the generation mode is similar to the candidate frequent pattern in the sequence in S31.5, and the details are not repeated herein;
s32.5: finding out a position list of two frequent patterns in the candidate frequent patterns between sequences with the length of k through the position list of the candidate frequent patterns in the sequences, generating a new position list by utilizing the position lists of the two frequent patterns, judging whether the length of the new position list is between the minimum support degree min _ s and the maximum support degree count max _ s, and adding the candidate frequent patterns between sequences with the length of k into the frequent pattern set between sequences if the length of the new position list is between the minimum support degree min _ s and the maximum support degree count max _ s; otherwise, deleting the candidate frequent pattern with the length of k among the sequences;
s32.6: judging whether the length of the frequent pattern in the inter-sequence frequent pattern set reaches a set maximum length maxBlocks, if not, repeatedly executing S32.4 and S32.5 until the length of the frequent pattern in the inter-sequence frequent pattern set reaches the set maximum length maxBlocks; and if the maximum length maxBlocks is reached, obtaining all the inter-sequence candidate frequent patterns of the satellite parameter data set.
In the above description, the time interval between two frequent patterns needs to be considered when mining the frequent patterns, and if the time interval is too long, the concept of "change in short time" cannot be embodied. Therefore, the embodiment of the present application determines the time interval t according to the starting positions s1, s2 of the two frequent patterns, i.e., t = s2-s 1. The range of t set in the embodiment of the present application is t e [1,4 ].
S40: generating association rules of the satellite parameter sequences according to the excavated inter-sequence frequent pattern, screening the generated association rules according to the set minimum confidence coefficient to obtain final association rules, and predicting the variation trend of the satellite parameters within a certain time in the future according to the final association rules;
the association rule generation mode is specifically as follows:
dividing the inter-sequence frequent pattern into two frequent patterns, namely a source parameter frequent pattern source and a target parameter frequent pattern target, and separating the divided last inter-sequence frequent pattern to be used as a quasi-target parameter frequent pattern freq;
calculating the confidence coefficient of the frequent patterns among the sequences, wherein the calculation formula of the confidence coefficient is as follows: conf = freq support count/source support count; where the support count is the length of the list of positions (i.e., how many times there are occurrences) of the inter-sequence frequent pattern.
And after the calculation is finished, judging whether conf is greater than the set minimum confidence coefficient min _ conf, if so, indicating that the frequent pattern among the sequences meets the requirement, and outputting the frequent pattern in an association rule form of source- > target.
The finally generated association rule is in the form of A (T) -B, and represents that the B event is likely to occur within T time after the A event occurs. The generated association rule is shown in fig. 5. Taking the first rule in the figure as an example, the rule indicates that when the satellite parameter sequence S0 has a normal trend of normal falling, normal rising and constant level, the satellite parameter sequence S3 has a 95% probability of having a trend of normal rising and normal rising in a short time.
Based on the above, the satellite time sequence parameter analysis method provided by the embodiment of the application carries out preprocessing on the satellite parameter sequence, converts the satellite parameter sequence into a data format capable of carrying out association rule analysis, carries out knowledge mining on the satellite parameter sequence by adopting an association relation analysis algorithm to obtain the inter-sequence frequent patterns of all the satellite parameter sequences, carries out association rule mining on the satellite parameter sequence according to the inter-sequence frequent patterns, and predicts the change trend of the satellite parameters within a certain time in the future according to the association rules. The invention can analyze the remote measurement parameters of a plurality of real-time satellites, solves the defect that the prior art cannot permit the multi-time sequence data of the on-orbit satellite, and provides reliable decision support for the safe and stable operation of the satellite.
Please refer to fig. 6, which is a schematic structural diagram of a data stream connection optimization system according to an embodiment of the present application. The data flow connection optimization system 40 according to the embodiment of the present application includes:
the data acquisition module 41: for obtaining a satellite parameter dataset;
the data mining module 42: the method comprises the steps of using an incidence relation analysis algorithm to conduct knowledge mining on a satellite parameter data set to obtain inter-sequence frequent patterns of all satellite timing sequence parameters in the satellite parameter data set;
the rule generation module 43: and the association rule is used for generating a satellite parameter data set according to the inter-sequence frequent pattern, and the change trend of the satellite parameters in the future set time is predicted according to the association rule.
Please refer to fig. 7, which is a schematic diagram of a terminal structure according to an embodiment of the present application. The terminal 50 comprises a processor 51, a memory 52 coupled to the processor 51.
The memory 52 stores program instructions for implementing the satellite timing parameter analysis method described above.
The processor 51 is operative to execute program instructions stored in the memory 52 to control data flow connection optimization.
The processor 51 may also be referred to as a CPU (Central Processing Unit). The processor 51 may be an integrated circuit chip having signal processing capabilities. The processor 51 may also be a general purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), an off-the-shelf programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Please refer to fig. 8, which is a schematic structural diagram of a storage medium according to an embodiment of the present application. The storage medium of the embodiment of the present application stores a program file 61 capable of implementing all the methods described above, where the program file 61 may be stored in the storage medium in the form of a software product, and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device) or a processor (processor) to execute all or part of the steps of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: various media capable of storing program codes, such as a usb disk, a mobile hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, or terminal devices, such as a computer, a server, a mobile phone, and a tablet.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method for analyzing satellite time sequence parameters is characterized by comprising the following steps:
acquiring a satellite parameter data set;
knowledge mining is carried out on the satellite parameter data set by adopting an incidence relation analysis algorithm to obtain a frequent pattern among sequences of all satellite timing sequence parameters in the satellite parameter data set;
and generating an association rule of the satellite parameter data set according to the inter-sequence frequent pattern, and predicting the change trend of the satellite parameters in the set time in the future according to the association rule.
2. The method of analyzing satellite timing parameters of claim 1, wherein said obtaining a satellite parameter data set further comprises:
and carrying out time alignment, data reduction and data discretization on the satellite parameter sequence in the satellite parameter data set, and converting the satellite parameter sequence into a set data format.
3. The method of claim 2, wherein the satellite timing parameters are analyzed,
the time alignment specifically comprises: deleting all satellite parameter sequences at a certain time point in the satellite parameter data set if one or more satellite parameter sequences are missing at the certain time point;
the data reduction specifically comprises the following steps: a sliding window w is given by using a segment aggregation approximation method, the mean value of all satellite parameter sequences in the current sliding window is calculated from the first satellite parameter sequence in the satellite parameter data set, and the mean value is used as the representative value of the satellite parameter sequences in the current sliding window; then, sliding backwards by a step length with the size of w, calculating the mean value of all satellite parameter sequences in the next sliding window, and taking the mean value as the representative value of the satellite parameter sequences in the next sliding window; repeating the steps until the calculation of all the satellite parameter sequences in the satellite parameter data set is completed;
the data discretization specifically comprises the following steps: capturing the variation trend between two adjacent satellite parameter sequences in the satellite parameter data set by adopting a monotonic characteristic extraction method, dividing the variation trend according to a set threshold value, and representing various variation trends through discretization symbols; the set thresholds comprise a normal _ ths threshold used for dividing the normal variation trend and an abrormal _ ths threshold used for dividing the abnormal variation trend; the normal change trend comprises constant level, normal rising and normal falling; the abnormal change trend comprises rapid rising and rapid falling.
4. The method according to any one of claims 1 to 3, wherein the performing knowledge mining on the satellite parameter data set by using an incidence relation analysis algorithm to obtain the inter-sequence frequent pattern of all the satellite timing parameters in the satellite parameter data set specifically comprises:
performing in-sequence frequent pattern mining on the satellite parameter data set to generate an in-sequence frequent pattern set of all satellite timing sequence parameters in the satellite parameter data set;
and mining inter-sequence frequent patterns of the satellite parameter data set according to the intra-sequence frequent pattern set to generate an inter-sequence frequent pattern set of all satellite timing sequence parameters in the satellite parameter data set.
5. The method according to claim 4, wherein the mining of inter-sequence frequent patterns from the satellite parameter data set according to the intra-sequence frequent pattern set specifically comprises:
given a minimum support count min _ s, a maximum support count max _ s, a maximum length windowsize:
selecting a frequent pattern IFPS _1 with the length of 1 in a sequence, wherein the IFPS _1 is composed of discretization symbols;
connecting every two discretization symbols in the in-sequence frequent pattern IFPS _1 to generate a candidate frequent pattern with the length of 2, scanning a satellite parameter data set, and finding and recording a position list of the candidate frequent pattern in the sequence; wherein, the position list records the starting position and the ending position of the candidate frequent pattern in each sequence;
judging whether the length of the position list of the candidate frequent patterns in the sequence is between the minimum support degree min _ s and the maximum support degree count max _ s, and if the length of the position list is between the minimum support degree min _ s and the maximum support degree count max _ s, adding the candidate frequent patterns in the sequence into a frequent pattern set in the sequence; otherwise, deleting the candidate frequent patterns in the sequence;
selecting two intra-sequence frequent patterns with the length of k-1, judging whether the sequence without the first discretization symbol of the frequent pattern in the first sequence is the same as the sequence without the last discretization symbol of the frequent pattern in the second sequence, and if the sequences are the same, connecting the discretization symbols of the intra-sequence frequent patterns in the two sequences to generate a candidate frequent pattern with the length of k in the sequence; wherein k is more than or equal to 2;
finding out position lists of two in-sequence frequent patterns in the candidate frequent patterns with the length of k, generating a new position list by using the position lists of the two in-sequence frequent patterns, judging whether the length of the new position list is between a minimum support degree count min _ s and a maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of k into a frequent pattern set in the sequence; otherwise, deleting the candidate frequent patterns with the length of k in the sequence;
and judging whether the length of the frequent pattern in the frequent pattern set in the sequence reaches a set maximum length windowsize, and if the length of the frequent pattern in the frequent pattern set in the sequence does not reach the set maximum length windowsize, regenerating a candidate frequent pattern in the sequence with the length of k until the length of the frequent pattern in the frequent pattern set in the sequence reaches the set maximum length windowsize.
6. The method according to claim 5, wherein the mining of inter-sequence frequent patterns from the satellite parameter data set according to the intra-sequence frequent pattern set specifically comprises:
giving an intra-sequence frequent pattern IFPS _1 and an inter-sequence frequent pattern maximum length maxBlocks; defining the position of a frequent pattern among sequences as the starting position of the first frequent pattern and the last frequent pattern in the frequent pattern set in the sequences;
connecting the frequent modes among different satellite parameter sequences in the intra-sequence frequent mode IFPS _1 pairwise from the intra-sequence frequent mode IFPS _1 to generate a candidate frequent mode with the length of 2 among the sequences;
generating a position list of the candidate frequent patterns with the length of 2 among the sequences through the position list of the candidate frequent patterns in the sequences, judging whether the length of the position list is between the minimum support degree count min _ s and the maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of 2 among the sequences into a frequent pattern set among the sequences; otherwise, deleting the inter-sequence candidate frequent pattern with the length of 2;
generating a candidate frequent pattern with the length of k among sequences by using two frequent patterns with the length of k-1;
finding out a position list of two frequent patterns in the candidate frequent patterns with the length of k from the position list of the candidate frequent patterns in the sequence, generating a new position list by using the position lists of the two frequent patterns, judging whether the length of the new position list is between the minimum support degree count min _ s and the maximum support degree count max _ s, and if so, adding the candidate frequent patterns with the length of k into the inter-sequence frequent pattern set; otherwise, deleting the candidate frequent patterns with the length of k among the sequences;
and judging whether the length of the frequent pattern in the inter-sequence frequent pattern set reaches a set maximum length maxBlock, and if not, regenerating the candidate frequent pattern with the length of k among the sequences until the length of the frequent pattern in the inter-sequence frequent pattern set reaches the set maximum length maxBlock.
7. The method according to claim 6, wherein the association rule for generating the satellite parameter data set according to the inter-sequence frequent pattern specifically comprises:
dividing the inter-sequence frequent pattern into a source parameter frequent pattern source and a target parameter frequent pattern target, and separating the divided last inter-sequence frequent pattern to be used as a quasi-target parameter frequent pattern freq;
calculating confidence of the inter-sequence frequent pattern: conf = freq support count/source support count; wherein the support count is a location list length of the inter-sequence frequent pattern;
and judging whether the confidence conf is greater than the set minimum confidence min _ conf, and if so, outputting the inter-sequence frequent pattern into an association rule form of source- > target.
8. A satellite timing parameter analysis system, comprising:
a data acquisition module: for obtaining a satellite parameter dataset;
a data mining module: the satellite parameter data set is subjected to knowledge mining by adopting an incidence relation analysis algorithm to obtain a sequence-to-sequence frequent pattern of all satellite timing sequence parameters in the satellite parameter data set;
a rule generation module: and the association rule is used for generating the satellite parameter data set according to the inter-sequence frequent pattern, and predicting the change trend of the satellite parameters in the set time in the future according to the association rule.
9. A terminal, comprising a processor, a memory coupled to the processor, wherein,
the memory stores program instructions for implementing the satellite timing parameter analysis method of any one of claims 1-7;
the processor is configured to execute the program instructions stored by the memory to control data flow connection optimization.
10. A storage medium having stored thereon program instructions executable by a processor to perform the method of satellite timing parameter analysis of any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110834911.4A CN113282645A (en) | 2021-07-23 | 2021-07-23 | Satellite time sequence parameter analysis method, system, terminal and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110834911.4A CN113282645A (en) | 2021-07-23 | 2021-07-23 | Satellite time sequence parameter analysis method, system, terminal and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113282645A true CN113282645A (en) | 2021-08-20 |
Family
ID=77287220
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110834911.4A Pending CN113282645A (en) | 2021-07-23 | 2021-07-23 | Satellite time sequence parameter analysis method, system, terminal and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113282645A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113821934A (en) * | 2021-09-30 | 2021-12-21 | 国网青海省电力公司电力科学研究院 | Method, device and equipment for predicting working condition parameters and storage medium |
| CN114630362A (en) * | 2022-03-16 | 2022-06-14 | 军事科学院系统工程研究院网络信息研究所 | Method for calculating support degree of space spectrum resources to satellite communication system downlink |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655857A (en) * | 2009-09-18 | 2010-02-24 | 西安建筑科技大学 | Method for mining data in construction regulation field based on associative regulation mining technology |
| CN106446081A (en) * | 2016-09-09 | 2017-02-22 | 西安交通大学 | Method for mining association relationship of time series data based on change consistency |
| US20170109653A1 (en) * | 2015-10-17 | 2017-04-20 | Tata Consultancy Services Limited | Multi-sensor data summarization |
| US20170124581A1 (en) * | 2011-02-28 | 2017-05-04 | Flytxt BV | Methods and systems for providing multivariate time series clustering for customer segmentation |
| CN107562865A (en) * | 2017-08-30 | 2018-01-09 | 哈尔滨工业大学深圳研究生院 | Multivariate time series association rule mining method based on Eclat |
| CN111046084A (en) * | 2019-12-18 | 2020-04-21 | 重庆大学 | Association rule mining method for multivariate time series monitoring data |
| CN112905689A (en) * | 2021-03-23 | 2021-06-04 | 河北工业大学 | Order-preserving sequence rule mining method |
-
2021
- 2021-07-23 CN CN202110834911.4A patent/CN113282645A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101655857A (en) * | 2009-09-18 | 2010-02-24 | 西安建筑科技大学 | Method for mining data in construction regulation field based on associative regulation mining technology |
| US20170124581A1 (en) * | 2011-02-28 | 2017-05-04 | Flytxt BV | Methods and systems for providing multivariate time series clustering for customer segmentation |
| US20170109653A1 (en) * | 2015-10-17 | 2017-04-20 | Tata Consultancy Services Limited | Multi-sensor data summarization |
| CN106446081A (en) * | 2016-09-09 | 2017-02-22 | 西安交通大学 | Method for mining association relationship of time series data based on change consistency |
| CN107562865A (en) * | 2017-08-30 | 2018-01-09 | 哈尔滨工业大学深圳研究生院 | Multivariate time series association rule mining method based on Eclat |
| CN111046084A (en) * | 2019-12-18 | 2020-04-21 | 重庆大学 | Association rule mining method for multivariate time series monitoring data |
| CN112905689A (en) * | 2021-03-23 | 2021-06-04 | 河北工业大学 | Order-preserving sequence rule mining method |
Non-Patent Citations (3)
| Title |
|---|
| 《干旱气象》: "数据挖掘技术在气象预报研究中的应用", 《干旱气象》 * |
| 赵益: "多时间序列上时序关联规则的挖掘", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 阿布•埃拉•哈桑尼: "《复杂系统中大数据分析与实践》", 31 August 2018, 《国防工业出版社》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113821934A (en) * | 2021-09-30 | 2021-12-21 | 国网青海省电力公司电力科学研究院 | Method, device and equipment for predicting working condition parameters and storage medium |
| CN113821934B (en) * | 2021-09-30 | 2024-01-19 | 国网青海省电力公司电力科学研究院 | Working condition parameter prediction method, device, equipment and storage medium |
| CN114630362A (en) * | 2022-03-16 | 2022-06-14 | 军事科学院系统工程研究院网络信息研究所 | Method for calculating support degree of space spectrum resources to satellite communication system downlink |
| CN114630362B (en) * | 2022-03-16 | 2022-10-21 | 军事科学院系统工程研究院网络信息研究所 | Method for calculating support degree of space spectrum resources to satellite communication system downlink |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113282645A (en) | Satellite time sequence parameter analysis method, system, terminal and storage medium | |
| JP2013206187A (en) | Information conversion device, information search device, information conversion method, information search method, information conversion program and information search program | |
| CN102985917B (en) | The generalized symbol of the similarity between uncertain time sequence represents | |
| US20210304364A1 (en) | Method and system for removing noise in documents for image processing | |
| KR20200052401A (en) | System Anomaly Behavior Analysis Technology based on Deep Learning Using Imaged Data | |
| US20200380409A1 (en) | Apparatus and method for analyzing time-series data based on machine learning | |
| CN112379859A (en) | Binary sampling processing method and device and countermeasure sample generating method and device | |
| CN101789082A (en) | Video identification | |
| CN112954455B (en) | Subtitle tracking method and device and electronic equipment | |
| CN113791755A (en) | Quantum random number sequence generation method, system and device | |
| CN111049684B (en) | Data analysis method, device, equipment and storage medium | |
| CN113886821A (en) | Malicious process identification method and device based on twin network, electronic equipment and storage medium | |
| Wei et al. | Generative steganography diffusion | |
| CN115567305B (en) | Sequential network attack prediction analysis method based on deep learning | |
| CN111552842A (en) | Data processing method, device and storage medium | |
| US20200133930A1 (en) | Information processing method, information processing system, and non-transitory computer readable storage medium | |
| CN113312619B (en) | Malicious process detection method and device based on small sample learning, electronic equipment and storage medium | |
| EP2953266B1 (en) | Data compression device, data compression method, and program | |
| CN108846083B (en) | Frequent pattern mining method and device | |
| CN109858977B (en) | User behavior analysis method and system based on self-coding | |
| CN114357849A (en) | User behavior abnormity detection method, system and terminal equipment | |
| EP4242881A1 (en) | Online detection of anomalous subsequences in time series | |
| CN117615170A (en) | Scene transition detection method and device, electronic equipment and storage medium | |
| KR101975463B1 (en) | Method and Apparatus for Integrated Processing Intelligent Situation Awareness | |
| WO2023139760A1 (en) | Data augmentation device, data augmentation method, and non-transitory computer-readable medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210820 |