CN107315777A - A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms - Google Patents
A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms Download PDFInfo
- Publication number
- CN107315777A CN107315777A CN201710402078.XA CN201710402078A CN107315777A CN 107315777 A CN107315777 A CN 107315777A CN 201710402078 A CN201710402078 A CN 201710402078A CN 107315777 A CN107315777 A CN 107315777A
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- information document
- monitoring information
- monitoring
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/903—Querying
- G06F16/90335—Query processing
- G06F16/90344—Query processing by using string matching techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/216—Parsing using statistical methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
-
- 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
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms.The present invention uses and is acquired sampling to power system monitor signal；Word segmentation processing and statistics are carried out to the monitoring signal read using the segmenting method based on string matching；Weight is calculated using TF IDF formula, spatial signature vectors are formed；Clustering is carried out to the sample set of spatial signature vectors using K nearest neighbor algorithms.Classified compression is carried out to the monitoring signal collected so as to realize.The present invention can effectively reduce monitoring alarm information content, compress signal of sending out and take place frequently by mistake, specification maintenance warning information, reduce until eliminating monitoring information brush screen phenomenon, prevent signal leakage prison, make the monitoring work to power network of monitoring personnel more easily and effectively rate, so as to improve the security of monitoring work, the operating pressure of monitoring personnel is reduced, is effectively prevented from omitting important power network abnormal signal, improve the security of power generation, it is ensured that the operation of electricity net safety stable.
Description
Technical field
The present invention relates to a kind of classified compression method of power system monitor signal, and in particular to one kind is based on K nearest neighbor algorithms
Power system monitor signal classified compression method.
Background technology
With the deep propulsion of Guo Wang companies " big operation " System Construction, Centralized Monitoring function turns into restriction power network development
Key issue.
Big, the low present situation of monitoring efficiency for monitoring signal amount, current domestic Surveillance center at different levels are substantially all according to country
Electricity grid substation Centralized Monitoring regulations for information administration, monitoring information is uniformly fallen into 5 types, the warning message on Centralized Monitoring interface
It can classify and centralized displaying.But do not improve in signal total number.Monitor's daily requirement is faced into large quantities
Stand the accident of end upload, abnormal, out-of-limit, displacement signal, power system monitor task is increasingly heavy, and monitoring work efficiency has to be hoisted.
In addition, a large amount of frequent movement, slot signals, easily cause monitoring alarm window brush screen, normal power system monitor work is had a strong impact on,
Even omit important power network abnormal signal.Based on above reality, it is necessary to optimize improvement to monitoring signal, reduction monitoring
The monitoring pressure of personnel, improves monitoring efficiency.
The content of the invention
Power system monitor signal is optimized in view of the above-mentioned problems, present invention offer is a kind of, the work of monitoring personnel is reduced
Pressure, improves the classified compression method of the power system monitor signal based on K nearest neighbor algorithms of monitoring personnel operating efficiency.
Because power system monitor informational capacity is big, monitor's daily requirement face into large quantities the accident of station end upload, exception,
Out-of-limit, displacement signal, power system monitor task is increasingly heavy.Work for convenience of monitor to monitoring power network, the present invention is proposed
Realized based on K nearest neighbor algorithms and monitoring signal is carried out the method for classified compression to solve this problem.
The characteristics of K nearest neighbor algorithms have simplicity, validity, be the best sorting algorithm based on vector space model it
One.
The step of classified compression is carried out to monitoring signal is implemented the following is based on K nearest neighbor algorithms：
Step 1: being acquired sampling to power system monitor signal, the power system monitor for including N number of monitoring information document T is set up
Message sample；
Step 2: using information sample of the segmenting method based on string matching to power system monitor, i.e. text training set S enters
Row word segmentation processing and statistics；
Step 3: calculating weight using TF-IDF formula, the two dimension of t monitoring information document T in text training set S is formed
Vectorial Dt, wherein 0 ＜ t≤N；All monitoring information document T bivector D in text training set StConstitute text training set S's
Spatial signature vectors；
Step 4: the bivector D of the alarm signal newly received to three calculating power system monitor backstages by step one；
Step 5: carrying out clustering to text training set S spatial signature vectors using K nearest neighbor algorithms；Pass through meter
Calculate the bivector D and text training set S of the alarm signal newly received monitoring information document T bivector DtSimilarity
The alarm signal newly received carries out classified compression in all kinds of monitoring information document T degree of membership to alarm signal.
The detailed process that the message sample of power system monitor is set up in step one is as follows：
The monitoring backstage that power network is sent in an item is produced when live grid equipment failure or displacement, an alarm is read
Information saves as a data sample that can be calculated, and power system monitor warning information is characterized as into formula form
T=[m, n] (1)
In formula:T is that the number of words n that warning information document, the m read is correspondence warning information is in warning information text
Hold, the message sample of power system monitor is N number of monitoring information document T set.
It is that text training set S carries out word segmentation processing and the detailed process counted to the message sample of power system monitor in step 2
It is as follows：
Can all there is space that Chinese character is carried out in each backlog information of generation separated, use space for cut-point
Monitoring information document T is divided into n entry, monitoring information document T characteristic item set is ultimately formed
θ={ δ1,δ2,δ3,...δn} (2),
Wherein δ is keyword that obtained entry after screening is monitoring information document T.
The step 3 Chinese version training set S specific forming process of spatial signature vectors is as follows：
Weight size of the entry in monitoring information document T is calculated using TF-IDF formula, its calculation formula is
In formula:f(δi, T) and it is entry δiThe frequency of occurrences in monitoring information document T；N is the individual of all monitoring information document T
Number；niTo contain entry δiMonitoring information document T number；Obtained weight coefficient is calculated by TF-IDF formula and characterizes word
The specificity of bar, ωi(T) it is more big, show entry δiSpecific higher, the ratio occurred in different monitoring information document T
It is lower, but the frequency occurred in single monitoring information document T is higher, then entry δiFor characterizing monitoring information document T's
Confidence level is higher；
Thus, it is possible to calculate the weight after the weight for obtaining different entries in monitoring information document T, normalized and be
WiAs entry δiWeight in monitoring information document T, therefore any monitoring information document T can be characterized as one
Individual bivector, its form is shown below { [δ1,W1],[δ2,W2].....[δn,Wn], if regarding different entries as one
Individual reference axis, then the bivector can be regarded as owning in a vector in the space, the message sample of power system monitor
Monitoring information document T bivector DtConstitute text training set S spatial signature vectors.
The detailed process for carrying out classified compression to the monitoring information document T read in step 5 is as follows：
Characteristic vector Dt={ Xt1,Xt2...Xtn}T{ X in 0 ＜ t≤Nt1,Xt2....XtnIt is text training set S
T-th of monitoring information document T bivector { [δt1,Wt1],[δt2,Wt2]..[δtn,Wtn]}；
The space that the bivector D and text training set S of text to be sorted are calculated using co sinus vector included angle formula is special
Levy the D of vectortSimilarity, formula is
The entry number that wherein n is included by monitoring information document T；
The k text most like with text to be sorted is selected as the arest neighbors of text to be sorted, according to k arest neighbors,
All monitoring information document T are divided into M classes in the message sample of power system monitor, calculate the bivector D of text to be sorted each
Individual classification TmIn degree of membership, wherein 0 ＜ m≤M；Calculation formula is
In formula：δ(D,Tm) represent if whether text to be sorted belongs to monitoring information document classification Tm, it is then 1 to be, otherwise
For 0, δ (D, Tm) calculation formula be
Select the maximum classification T of degree of membershipm, text to be sorted is not included into category TmIn, so as to realize to power network
The backlog information that equipment is produced carries out classified compression.
The present invention can effectively reduce monitoring alarm information content, compress signal of sending out and take place frequently by mistake, specification maintenance alarm letter
Breath, is reduced until elimination monitoring information brush screen phenomenon, prevents signal leakage prison, make the monitoring work to power network of monitoring personnel more
Easily and effectively rate, so as to improve the security of monitoring work, reduces the operating pressure of monitoring personnel, is effectively prevented from omitting important
Power network abnormal signal, improves the security of power generation, it is ensured that the operation of electricity net safety stable.
Embodiment
A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms, comprises the following steps：
Step 1: being acquired sampling to power system monitor signal, the power system monitor for including N number of monitoring information document T is set up
Message sample；
The monitoring backstage that power network is sent in an item is produced when live grid equipment failure or displacement, an alarm is read
Information saves as a data sample that can be calculated, and power system monitor warning information is characterized as into formula form
T=[m, n] (1)
In formula:T is that the number of words n that warning information document, the m read is correspondence warning information is in warning information text
Hold, the message sample of power system monitor is N number of monitoring information document T set.
Step 2: being text training set S to the message sample of power system monitor using the segmenting method based on string matching
Carry out word segmentation processing and statistics；
Can all there is space that Chinese character is carried out in each backlog information of generation separated, use space for cut-point
Monitoring information document T is divided into n entry, monitoring information document T characteristic item set is ultimately formed
θ={ δ1,δ2,δ3,...δn} (2),
Wherein δ is keyword that obtained entry after screening is monitoring information document T.
Step 3: calculating weight using TF-IDF formula, the two dimension of t monitoring information document T in text training set S is formed
Vectorial Dt, wherein 0 ＜ t≤N；All monitoring information document T bivector D in text training set StConstitute text training set S's
Spatial signature vectors；
Weight size of the entry in monitoring information document T is calculated using TF-IDF formula, its calculation formula is
In formula:f(δi, T) and it is entry δiThe frequency of occurrences in monitoring information document T；N is the individual of all monitoring information document T
Number；niTo contain entry δiMonitoring information document T number；Obtained weight coefficient is calculated by TF-IDF formula and characterizes word
The specificity of bar, ωi(T) it is more big, show entry δiSpecific higher, the ratio occurred in different monitoring information document T
It is lower, but the frequency occurred in single monitoring information document T is higher, then entry δiFor characterizing monitoring information document T's
Confidence level is higher；
Thus, it is possible to calculate the weight after the weight for obtaining different entries in monitoring information document T, normalized and be
WiAs entry δiWeight in monitoring information document T, therefore any monitoring information document T can be characterized as one
Individual bivector, its form is shown below { [δ1,W1],[δ2,W2].....[δn,Wn], if regarding different entries as one
Individual reference axis, then the bivector can be regarded as owning in a vector in the space, the message sample of power system monitor
Monitoring information document T bivector DtThe space for constituting text training set S is special
Step 4: the bivector D of the alarm signal newly received to three calculating power system monitor backstages by step one；
Step 5: carrying out clustering to text training set S spatial signature vectors using K nearest neighbor algorithms；Pass through meter
Calculate the bivector D and text training set S of the alarm signal newly received monitoring information document T bivector DtSimilarity
The alarm signal newly received carries out classified compression in all kinds of monitoring information document T degree of membership to alarm signal.
Characteristic vector Dt={ Xt1,Xt2...Xtn}T{ X in 0 ＜ t≤Nt1,Xt2....XtnIt is text training set S
T-th of monitoring information document T bivector { [δt1,Wt1],[δt2,Wt2]..[δtn,Wtn]}；
The space that the bivector D and text training set S of text to be sorted are calculated using co sinus vector included angle formula is special
Levy the D of vectortSimilarity, formula is
The entry number that wherein n is included by monitoring information document T；
The k text most like with text to be sorted is selected as the arest neighbors of text to be sorted, according to k arest neighbors,
All monitoring information document T are divided into M classes in the message sample of power system monitor, calculate the bivector D of text to be sorted each
Individual classification TmIn degree of membership, wherein 0 ＜ m≤M；Calculation formula is
In formula：δ(D,Tm) represent if whether text to be sorted belongs to monitoring information document classification Tm, it is then 1 to be, otherwise
For 0, δ (D, Tm) calculation formula be
Select the maximum classification T of degree of membershipm, text to be sorted is not included into category TmIn, so as to realize to power network
The backlog information that equipment is produced carries out classified compression.
Claims (5)
1. a kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms, it is characterised in that comprise the following steps：
Step 1: being acquired sampling to power system monitor signal, the letter of the power system monitor comprising N number of monitoring information document T is set up
Cease sample；
Step 2: being that text training set S is divided to the information sample of power system monitor using the segmenting method based on string matching
Word processing and statistics；
Step 3: calculating weight using TF-IDF formula, the bivector of t monitoring information document T in text training set S is formed
Dt, wherein 0 ＜ t≤N；All monitoring information document T bivector D in text training set StConstitute text training set S space
Characteristic vector；
Step 4: the bivector D of the alarm signal newly received to three calculating power system monitor backstages by step one；
Step 5: carrying out clustering to text training set S spatial signature vectors using K nearest neighbor algorithms；It is new by calculating
The bivector D and text training set S of the alarm signal received monitoring information document T bivector DtSimilarity and new
The alarm signal received carries out classified compression in all kinds of monitoring information document T degree of membership to alarm signal.
2. the classified compression method of the power system monitor signal according to claim 1 based on K nearest neighbor algorithms, its feature exists
The detailed process that the message sample of power system monitor is set up in step one is as follows：
The monitoring backstage that power network is sent in an item is produced when live grid equipment failure or displacement, a warning information is read
A data sample that can be calculated is saved as, power system monitor warning information is characterized as formula form
T=[m, n] (1)
In formula:T is the number of words that warning information document, the m read is correspondence warning information, and n is warning information content of text, electricity
The set that the message sample of net monitoring is N number of monitoring information document T.
3. the classified compression method of the power system monitor signal according to claim 2 based on K nearest neighbor algorithms, its feature exists
Message sample to power system monitor in step 2 is that the detailed process that text training set S carries out word segmentation processing and counted is as follows：
Can all there is space that Chinese character is carried out in each backlog information of generation separated, it is that cut-point handle is supervised to use space
Control information document T is divided into n entry, ultimately forms monitoring information document T characteristic item set θ={ δ1,δ2,δ3,...δn}
(2), wherein δ is monitoring information document T keyword for obtained entry after screening.
4. the classified compression method of the power system monitor signal according to claim 3 based on K nearest neighbor algorithms, its feature exists
It is as follows in the step 3 Chinese version training set S specific forming process of spatial signature vectors：
Weight size of the entry in monitoring information document T is calculated using TF-IDF formula, its calculation formula is
<mrow>
<msub>
<mi>&omega;</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>T</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&delta;</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mi>T</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>l</mi>
<mi>o</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mi>N</mi>
<msub>
<mi>n</mi>
<mi>i</mi>
</msub>
</mfrac>
<mo>+</mo>
<mn>0.01</mn>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula:f(δi, T) and it is entry δiThe frequency of occurrences in monitoring information document T；N is all monitoring information document T number；
niTo contain entry δiMonitoring information document T number；Obtained weight coefficient is calculated by TF-IDF formula and characterizes entry
Specificity, ωi(T) it is more big, show entry δiSpecificity it is higher, the ratio occurred in different monitoring information document T is lower,
But the frequency occurred in single monitoring information document T is higher, then entry δiConfidence level for characterizing monitoring information document T
It is higher；
Thus, it is possible to calculate the weight after the weight for obtaining different entries in monitoring information document T, normalized and be
<mrow>
<msub>
<mi>W</mi>
<mi>i</mi>
</msub>
<mo>=</mo>
<mfrac>
<mrow>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&delta;</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mi>T</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>l</mi>
<mi>o</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mi>N</mi>
<msub>
<mi>n</mi>
<mi>i</mi>
</msub>
</mfrac>
<mo>+</mo>
<mn>0.01</mn>
<mo>)</mo>
</mrow>
</mrow>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>&delta;</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<mi>T</mi>
<mo>)</mo>
</mrow>
<mo>&times;</mo>
<mi>l</mi>
<mi>o</mi>
<mi>g</mi>
<mrow>
<mo>(</mo>
<mfrac>
<mi>N</mi>
<msub>
<mi>n</mi>
<mi>i</mi>
</msub>
</mfrac>
<mo>+</mo>
<mn>0.01</mn>
<mo>)</mo>
</mrow>
</mrow>
</msqrt>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>4</mn>
<mo>)</mo>
</mrow>
</mrow>
WiAs entry δiWeight in monitoring information document T, therefore any monitoring information document T can be characterized as one two
Dimensional vector, its form is shown below { [δ1,W1],[δ2,W2].....[δn,Wn], if regarding different entries as a seat
Parameter, then the bivector can be regarded as all monitoring in a vector in the space, the message sample of power system monitor
Information document T bivector DtConstitute text training set S spatial signature vectors.
5. the classified compression method of the power system monitor signal according to claim 4 based on K nearest neighbor algorithms, its feature exists
The detailed process for carrying out classified compression to the monitoring information document T read in step 5 is as follows：
Characteristic vector Dt={ Xt1,Xt2...Xtn}T{ X in 0 ＜ t≤Nt1,Xt2....XtnIt is t-th of text training set S
Monitoring information document T bivector
{[δt1,Wt1],[δt2,Wt2]..[δtn,Wtn]}；
Calculated using co sinus vector included angle formula text to be sorted bivector D and text training set S space characteristics to
The D of amounttSimilarity, formula is
<mrow>
<mi>s</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mo>,</mo>
<msub>
<mi>D</mi>
<mi>t</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mo>&CenterDot;</mo>
<msub>
<mi>X</mi>
<mrow>
<mi>t</mi>
<mi>i</mi>
</mrow>
</msub>
</mrow>
<mrow>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msup>
<msub>
<mi>X</mi>
<mi>i</mi>
</msub>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
<mo>&CenterDot;</mo>
<msqrt>
<mrow>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msup>
<msub>
<mi>X</mi>
<mrow>
<mi>t</mi>
<mi>i</mi>
</mrow>
</msub>
<mn>2</mn>
</msup>
</mrow>
</msqrt>
</mrow>
</mfrac>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
The entry number that wherein n is included by monitoring information document T；
The k text most like with text to be sorted is selected as the arest neighbors of text to be sorted, according to k arest neighbors, power network
All monitoring information document T are divided into M classes in the message sample of monitoring, calculate the bivector D of text to be sorted in each class
Other TmIn degree of membership, wherein 0 ＜ m≤M；Calculation formula is
<mrow>
<mi>p</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>t</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
<mi>k</mi>
</munderover>
<mi>s</mi>
<mi>i</mi>
<mi>m</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mo>,</mo>
<msub>
<mi>D</mi>
<mi>t</mi>
</msub>
<mo>)</mo>
</mrow>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>,</mo>
<mn>0</mn>
<mo><</mo>
<mi>m</mi>
<mo>&le;</mo>
<mi>M</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula：δ(D,Tm) represent if whether text to be sorted belongs to monitoring information document classification Tm, it is then 1 to be, is otherwise 0, δ
(D,Tm) calculation formula be
<mrow>
<mi>&delta;</mi>
<mrow>
<mo>(</mo>
<mi>D</mi>
<mo>,</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfenced open = "{" close = "">
<mtable>
<mtr>
<mtd>
<mrow>
<mn>1</mn>
<mo>,</mo>
<mi>D</mi>
<mo>&Element;</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<mn>0</mn>
<mo>,</mo>
<mi>D</mi>
<mo>&NotElement;</mo>
<msub>
<mi>T</mi>
<mi>m</mi>
</msub>
</mrow>
</mtd>
</mtr>
</mtable>
</mfenced>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
</mrow>
Select the maximum classification T of degree of membershipm, text to be sorted is not included into category TmIn, so as to realize to grid equipment
The backlog information of generation carries out classified compression.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710402078.XA CN107315777A (en) | 2017-05-31 | 2017-05-31 | A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710402078.XA CN107315777A (en) | 2017-05-31 | 2017-05-31 | A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107315777A true CN107315777A (en) | 2017-11-03 |
Family
ID=60184016
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710402078.XA Pending CN107315777A (en) | 2017-05-31 | 2017-05-31 | A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107315777A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107947929A (en) * | 2017-12-28 | 2018-04-20 | 中南大学 | Continuous variable quantum key distribution system and implementation method based on k neighbours processing |
| CN108197109A (en) * | 2017-12-29 | 2018-06-22 | 北京百分点信息科技有限公司 | A kind of multilingual analysis method and device based on natural language processing |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574165A (en) * | 2015-12-17 | 2016-05-11 | 国家电网公司 | Power grid operation monitoring information identification and classification method based on clustering |
| CN106228280A (en) * | 2016-07-06 | 2016-12-14 | 吴本刚 | Grid operating monitoring information identification categorizing system |
-
2017
- 2017-05-31 CN CN201710402078.XA patent/CN107315777A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105574165A (en) * | 2015-12-17 | 2016-05-11 | 国家电网公司 | Power grid operation monitoring information identification and classification method based on clustering |
| CN106228280A (en) * | 2016-07-06 | 2016-12-14 | 吴本刚 | Grid operating monitoring information identification categorizing system |
Non-Patent Citations (2)
| Title |
|---|
| 余振华，杨世勇，舒征宇，李黄强，张伟奇: "基于聚类的电网运行监控信息辨识分类方法研究", 《湖北电力》 * |
| 张成松: "电网基建项目全面风险管理及其信息系统研究", 《中国博士学位论文全文数据库经济与管理科学辑》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107947929A (en) * | 2017-12-28 | 2018-04-20 | 中南大学 | Continuous variable quantum key distribution system and implementation method based on k neighbours processing |
| CN107947929B (en) * | 2017-12-28 | 2021-04-23 | 中南大学 | Continuous variable quantum key distribution system based on k-nearest neighbor processing and implementation method |
| CN108197109A (en) * | 2017-12-29 | 2018-06-22 | 北京百分点信息科技有限公司 | A kind of multilingual analysis method and device based on natural language processing |
| CN108197109B (en) * | 2017-12-29 | 2021-04-23 | 北京百分点科技集团股份有限公司 | Multi-language analysis method and device based on natural language processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106202886B (en) | Track circuit red band Fault Locating Method based on fuzzy coarse central and decision tree | |
| CN103581186B (en) | A kind of network security situational awareness method and system | |
| CN105574165B (en) | A kind of grid operating monitoring information identification classification method based on cluster | |
| CN102487293B (en) | Satellite communication network abnormity detection method based on network control | |
| CN108009690B (en) | Ground bus stealing group automatic detection method based on modularity optimization | |
| CN101718634B (en) | Equipment state comprehensive dynamic alarming method based on multivariate probability model | |
| CN102855492B (en) | Classification method based on mineral flotation foam image | |
| CN103077347B (en) | A kind of hybrid intrusion detection method based on improving the fusion of kernel vector machine data | |
| CN106203492A (en) | The system and method that a kind of image latent writing is analyzed | |
| CN110737976B (en) | Mechanical equipment health assessment method based on multidimensional information fusion | |
| CN107315777A (en) | A kind of classified compression method of the power system monitor signal based on K nearest neighbor algorithms | |
| CN114024829B (en) | Fault maintenance method, device, equipment and storage medium for power communication network | |
| CN109344765A (en) | A kind of intelligent analysis method entering shop personnel analysis for chain shops | |
| CN107483451A (en) | Based on serial parallel structural network secure data processing method and system, social networks | |
| CN111163075B (en) | Dynamic adjustment method for performance index threshold of power monitoring system equipment | |
| CN116011330A (en) | Power system voltage stability evaluation method based on improved random forest | |
| CN107992902A (en) | A kind of routine bus system based on supervised learning steals individual automatic testing method | |
| CN108280357A (en) | Data leakage prevention method, system based on semantic feature extraction | |
| CN105205312B (en) | The genetic analysis of road accident focus and damage degree appraisal procedure | |
| Chen et al. | Intrusion detection system based on immune algorithm and support vector machine in wireless sensor network | |
| CN110059189B (en) | Game platform message classification system and method | |
| CN106529837A (en) | Association method of wind data and transmission line body operation data | |
| CN104899606A (en) | Steganalysis method based on local learning | |
| CN115242607A (en) | Power network reliability early warning method based on node importance | |
| Ishwarya et al. | Seperation of phishing emails using probabilistic classifiers |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20171103 |
|
| RJ01 | Rejection of invention patent application after publication |