CN110309885A - Computer room state judging method based on big data - Google Patents
Computer room state judging method based on big data Download PDFInfo
- Publication number
- CN110309885A CN110309885A CN201910605879.5A CN201910605879A CN110309885A CN 110309885 A CN110309885 A CN 110309885A CN 201910605879 A CN201910605879 A CN 201910605879A CN 110309885 A CN110309885 A CN 110309885A
- Authority
- CN
- China
- Prior art keywords
- state
- computer room
- collection
- comparison
- picture
- 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
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Operations Research (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Quality & Reliability (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Strategic Management (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Alarm Systems (AREA)
Abstract
Computer room state judging method based on big data, is related to computer room detection technique field.The present invention is the abnormality monitoring method in order to solve existing computer room, be easy there is a problem of slip, can not find the problem in time and can not accurate judgement computer room state cause maintenance efficiency low.Computer room state judging method of the present invention based on big data acquires the abnormality picture of computer room by big data；It replicates multiple identical comparisons collection and is synchronized respectively with state set and is compared, so that comparing speed faster, can timely obtain current information, more efficiently, and the data of duplication be deleted again after comparison, reduce EMS memory occupation；Since state has carried out state classification first, then it is capable of the classification of direct lockout issue after comparing similarity；The state in computer room can be monitored in real time, it is more efficient, it finds the problem much sooner, the monitoring suitable for all computer rooms.
Description
Technical field
The invention belongs to computer room detection technique fields.
Background technique
Power scheduling is to guarantee that power network safety operation, external reliable power supply, the work of all kinds of power generations are orderly
A kind of effective management means for carrying out and using.The specific works content of power scheduling is anti-according to various information acquisition equipment
Be fed back to the data information come or information that monitoring personnel provides, in conjunction with power grid actual operation parameters, as voltage, electric current, frequency,
Load etc. comprehensively considers every production work development condition, judges power grid security, economical operation state, passes through phone
Or automatic system issues operational order, commander site operation personnel or automatic control system are adjusted, and are such as adjusted generator and are gone out
Power, Load adjustment distribution, switched capacitor, reactor etc., so that it is guaranteed that power grid continues safe and stable operation.Therefore, electric power tune
Degree computer room needs to guarantee at runtime safety and steady, otherwise once breaking down, it will leads to power-off, load increase, voltage not
The problems such as stablizing.
It is at present video monitoring or sensor alarm to the abnormality monitoring method of power scheduling computer room.In computer room, example
Such as the problems such as temperature, humidity, smog dust, corresponding sensor real-time monitoring can be used and obtain.But as ageing equipment or
The problems such as fracture is then cannot be directly detected by sensor.For problems, a kind of mode is at supervision object
Monitoring of the video camera realization to equipment and its operating condition is installed.But artificial monitoring requirement staff is always before screen,
Not only heavy workload, and inevitably have careless omission.Another way is that staff periodically carries out inspection, and regularly checks not
Can play the role of real time monitoring, cause it is problematic can not find in time, just have such as route loosen, aging or attachment
Dust causes device temperature to increase, generates the generation of phenomena such as smog in turn.Although these phenomenons can pass through sensing after occurring
Device collects corresponding signal, and still, sensor can only reflect the presentation of problem after deterioration, can not inform staff's problem
It is basic, staff does not know the time of day of computer room, it is necessary to first check problem, can waste the plenty of time at this time.
To sum up, for the security monitoring of power scheduling computer room, there is also many masty problems at present.
Summary of the invention
The present invention is the abnormality monitoring method in order to solve existing computer room, is easy to have careless omission, can not find to ask in time
Topic and can not the accurate judgement computer room state problem that causes maintenance efficiency low, the computer room state based on big data is now provided and is sentenced
Disconnected method.
Computer room state judging method based on big data, comprising the following steps:
Step 1: establishing property data base using network big data, includes the abnormal shape of n type in this feature database
State collection and 1 normal condition collection；
Step 2: the picture of acquisition computer room position to be judged, the feature for extracting the picture collect as a comparison and replicate n again
A identical comparison collection；
Step 3: n+1 comparison collection is corresponded with n+1 state set in property data base respectively, will be corresponding
Comparison collection and state set as a control group, calculate the similarity of comparison collection and state set in each control group, obtain n+1
Then a similarity result deletes the comparison collection of n duplication；
Step 4: judge whether maximum value is greater than 50% in n+1 similarity result, be to then follow the steps five, otherwise hold
Row step 6；
Step 5: by the group as a result of control group corresponding to similarity maximum value, by the state of state set in result group
As computer room state；
Step 6: the picture that step 2 obtains is sent to computer room control centre, machine is judged according to picture by technical staff
Room state.
Step 7: whether the judging result that judgment step six obtains belongs to state corresponding to normal condition collection, then will be
The element deposit normal condition that comparison is concentrated is concentrated, and step 8 is otherwise executed；
Step 8: whether the judging result that judgment step six obtains belongs to state corresponding to n abnormality collection, is then
The element that comparison is concentrated is stored in corresponding abnormality and concentrates, and otherwise executes step 9；
Step 9: comparison is collected into the abnormality collection new as one and is stored in property data base.
Step 5 using the state of state set in result group as computer room state after, return step two is to subsequent time
Computer room state is judged；
The element that normal condition collection and step 8 concentrate comparison, which is stored in, in the element that step 7 concentrates comparison is stored in phase
After the abnormality collection answered, property data base is updated, then computer room state of the return step two to subsequent time
Judged；
After step 9, property data base is updated, makes n=n+1, then computer room shape of the return step two to subsequent time
State is judged.
In step 3, comparison collection and the similarity of state set in each control group are calculated method particularly includes:
Each element that comparison is concentrated is compared with each element in state set respectively and obtains the phase between element
Like angle value, similarity value maximum between element is collected and the similarity of state set as a comparison.
In step 1, property data base is established method particularly includes:
The picture in computer room under various abnormalities in position to be judged is acquired in network large database concept, extracts each picture
Feature and using feature corresponding to same Exception Type state as an abnormality set, obtain the exception of n type
State set,
The picture in computer room under position normal condition to be judged is acquired in network large database concept, is acquired wait judge in computer room
Picture under position normal condition to be judged, extracts the picture feature under each normal condition, as normal condition set,
Property data base is established using n abnormality set and normal condition set.
In step 1,2 features are at least extracted to each picture.
After property data base is established in step 1, property data base is uploaded in block chain.
After property data base updates, updated property data base is uploaded in block chain.
Computer room state judging method of the present invention based on big data acquires the abnormality of computer room by big data
Picture refers to as a comparison, and information is more comprehensive, and comparing result is relatively reliable；Replicate multiple identical comparisons collection respectively with shape
State collection, which synchronizes, to be compared, so that comparing speed faster, can timely be obtained current information, more efficiently, and compared
The data of duplication are deleted again afterwards, reduce EMS memory occupation；Since state has carried out state classification first, then when comparison similarity it
It is capable of the classification of direct lockout issue afterwards, staff can learn the root of problem at the first time, to the subsequent of staff
Maintenance provides effective information auxiliary；Package can in real time be monitored the state in computer room, more efficient, discovery
Problem much sooner, the monitoring suitable for all computer rooms.
Detailed description of the invention
Fig. 1 is the flow chart of the computer room state judging method based on big data described in specific embodiment one.
Specific embodiment
Specific embodiment 1: present embodiment is illustrated referring to Fig.1, based on big data described in present embodiment
Computer room state judging method, comprising the following steps:
Step 1: the picture in computer room under various abnormalities in position to be judged is acquired in network large database concept, each
Picture at least extracts 2 features and using feature corresponding to same Exception Type state as an abnormality set, obtains
The abnormality set of n type；
The picture in computer room under position normal condition to be judged is acquired in network large database concept, is acquired wait judge in computer room
Picture under position normal condition to be judged, extracts the picture feature under each normal condition, as normal condition set；
Property data base is established using n abnormality set and normal condition set, property data base is uploaded to area
In block chain.
In the above step 1, material is acquired in network large database concept to refer to as a comparison, these network large database concepts
In data class multiplicity, to it is subsequent comparison and judgement provide more comprehensive basis.Also, in advance by Status Type point
Class convenient for subsequent contrast and determines type at collection.
Step 2: the picture of acquisition computer room position to be judged, the feature for extracting the picture collect as a comparison and replicate n again
A identical comparison collection.
Step 3: n+1 comparison collection is corresponded with n+1 state set in property data base respectively, will be corresponding
Comparison collection and state set as a control group；
In control group, each element that comparison is concentrated is compared and obtained with each element in state set respectively
Similarity value between element, similarity value maximum between element collect and the similarity of state set as a comparison, obtains n+1 altogether
The similarity result of a comparison collection and state set；
It has been determined that similarity result and then the comparison collection for replicating n are deleted.
In above-mentioned steps two, then comparison collection duplication is existed so that the quantity of comparison collected works is equal with the quantity of state set
Compared correspondingly in step 3.The mode that compared to one comparison collection is compared with multiple state sets respectively, section
About the time, improve work efficiency.And then replicated comparison collection is deleted, the occupancy of memory can be reduced, is saved empty
Between.Further, since the quantity of element is more in set, if an element of comparison concentration and an element phase in state set
Together, then the state consistency of position and state set to be judged is proved.But if an element and state set that comparison is concentrated
Entirety is compared, then similarity just will be greatly reduced, with for impossible and each state set element after all it is identical, because
This will lead to the problem of result inaccuracy.In step 3, element and element are compared, then maximum similarity value is made
For the similarity result between gathering and gathering, it is only so most true as a result, last judging result is also more accurate.
Step 4: judge whether maximum value is greater than 50% in n+1 similarity result, be to then follow the steps five, otherwise hold
Row step 6；
Step 5: by the group as a result of control group corresponding to similarity maximum value, by the state of state set in result group
As computer room state, then return step two judges the computer room state of subsequent time；
Step 6: the picture that step 2 obtains is sent to computer room control centre, machine is judged according to picture by technical staff
Room state.
In above-mentioned steps four, when the maximum value in similarity result is greater than 50%, then position to be judged can determine
It is the state of state set that state, which has more than half probability, then can by step 5 using the state of state set in result group as
Computer room state, to complete the state judgement at current time.
In above-mentioned steps four, when the maximum value in similarity result is less than 50%, then position to be judged can determine
State and the state of state set in feature database less meet, then just needing further to judge by step 6, it may be assumed that by technology
Personnel's artificial judgment.Although this step is also required to using artificial judgment, due to passed through in above-mentioned steps big data into
Go judgement, then just illustrate that the state of position to be judged is a kind of special state that probability of occurrence is extremely low or new state, by
Lack in what such state occurred, then the probability that step 6 is triggered also will be reduced accordingly, necessarily also reduces and manually sentence
Disconnected probability；And staff is just notified that when only state occurs, not needing staff's moment monitors, significantly
Reduce the workload of staff.
Step 7: whether the judging result that judgment step six obtains belongs to state corresponding to normal condition collection, then will be
The element deposit normal condition that comparison is concentrated is concentrated, and updates property data base, updated property data base is uploaded to block
In chain, return step two judges the computer room state of subsequent time, otherwise executes step 8；
Step 8: whether the judging result that judgment step six obtains belongs to state corresponding to n abnormality collection, is then
The element that comparison is concentrated is stored in corresponding abnormality and concentrates, and updates property data base, will be on updated property data base
It reaches in block chain, return step two judges the computer room state of subsequent time, otherwise executes step 9；
Step 9: comparison is collected into the abnormality collection new as one and is stored in property data base, property data base is updated
Make n=n+1, updated property data base is uploaded in block chain, then computer room shape of the return step two to subsequent time
State is judged.
After step 6 artificial judgment, it will be able to Status Type is determined, if the Status Type is state in property data base
The corresponding type of collection, element is just directly stored in corresponding state collection by that.If the Status Type is shape in property data base
Type corresponding to state collection, then prove new type occur, and the type necessarily Exception Type, then in property data base
One new abnormality collection of middle increase, so that n=n+1, property data base is updated, so that data therein more add
It is whole, improve subsequent judging nicety rate.
Moreover, after property data base obtains update, updated property data base is uploaded to block in above-mentioned steps
In chain, information sharing, once a certain system failure, can also find database in block chain, subsequent detection is protected.
Claims (8)
1. the computer room state judging method based on big data, which comprises the following steps:
Step 1: establishing property data base using network big data, includes the abnormality collection of n type in this feature database
With 1 normal condition collection；
Step 2: the picture of acquisition computer room position to be judged, the feature for extracting the picture collect as a comparison and replicate n phase again
Same comparison collection；
Step 3: n+1 comparison collection is corresponded with n+1 state set in property data base respectively, will be corresponding right
Than collection and state set as a control group, calculate the similarity of comparison collection and state set in each control group, obtain n+1 phase
Like degree as a result, then deleting the comparison collection of n duplication；
Step 4: judge whether maximum value is greater than 50% in n+1 similarity result, be to then follow the steps five, otherwise execute step
Rapid six；
Step 5: by the group as a result of control group corresponding to similarity maximum value, using the state of state set in result group as
Computer room state；
Step 6: the picture that step 2 obtains is sent to computer room control centre, computer room shape is judged according to picture by technical staff
State.
2. the computer room state judging method according to claim 1 based on big data, which is characterized in that further include following step
It is rapid:
Step 7: whether the judging result that judgment step six obtains belongs to state corresponding to normal condition collection, and being then will comparison
The element deposit normal condition of concentration is concentrated, and step 8 is otherwise executed；
Step 8: whether the judging result that judgment step six obtains belongs to state corresponding to n abnormality collection, and being then will be right
Corresponding abnormality is stored in than the element of concentration to concentrate, and otherwise executes step 9；
Step 9: comparison is collected into the abnormality collection new as one and is stored in property data base.
3. the computer room state judging method according to claim 2 based on big data, which is characterized in that
Step 5 using the state of state set in result group as computer room state after, computer room of the return step two to subsequent time
State is judged；
The element that the element deposit normal condition collection and step 8 of comparison concentration concentrate comparison is stored in accordingly in step 7
After abnormality collection, property data base is updated, then return step two carries out the computer room state of subsequent time
Judgement；
After step 9, update property data base, make n=n+1, then return step two to the computer room state of subsequent time into
Row judgement.
4. the computer room state judging method according to claim 1,2 or 3 based on big data, which is characterized in that step 3
In, calculate comparison collection and the similarity of state set in each control group method particularly includes:
Each element that comparison is concentrated is compared with each element in state set respectively and obtains the similarity between element
Value collects similarity value maximum between element and the similarity of state set as a comparison.
5. the computer room state judging method according to claim 4 based on big data, which is characterized in that in step 1, build
Vertical property data base method particularly includes:
The picture in computer room under various abnormalities in position to be judged is acquired in network large database concept, extracts the spy of each picture
It levies and using feature corresponding to same Exception Type state as an abnormality set, the abnormality of n type of acquisition
Set,
The picture in computer room under position normal condition to be judged is acquired in network large database concept, is acquired wait judge in computer room wait sentence
Picture under disconnected position normal condition, extracts the picture feature under each normal condition, as normal condition set,
Property data base is established using n abnormality set and normal condition set.
6. the computer room state judging method according to claim 5 based on big data, which is characterized in that right in step 1
Each picture at least extracts 2 features.
7. the computer room state judging method according to claim 5 based on big data, which is characterized in that feature in step 1
After Database, property data base is uploaded in block chain.
8. the computer room state judging method according to claim 7 based on big data, which is characterized in that property data base is more
After new, updated property data base is uploaded in block chain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910605879.5A CN110309885A (en) | 2019-07-05 | 2019-07-05 | Computer room state judging method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910605879.5A CN110309885A (en) | 2019-07-05 | 2019-07-05 | Computer room state judging method based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110309885A true CN110309885A (en) | 2019-10-08 |
Family
ID=68079240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910605879.5A Pending CN110309885A (en) | 2019-07-05 | 2019-07-05 | Computer room state judging method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110309885A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111209944A (en) * | 2019-12-31 | 2020-05-29 | 上海索广电子有限公司 | Self-adaptive image detection system and image detection method |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7512275B2 (en) * | 2003-10-21 | 2009-03-31 | Sharp Kabushiki Kaisha | Image collating apparatus, image collating method, image collating program and computer readable recording medium recording image collating program |
| CN101582195A (en) * | 2009-06-09 | 2009-11-18 | 深圳中兴力维技术有限公司 | Method for generating alarm in dynamic environment monitoring system |
| CN103218464A (en) * | 2013-03-23 | 2013-07-24 | 中国水利电力物资有限公司 | Wind turbine generator state-based wind turbine generator data storage method and system |
| CN105116872A (en) * | 2015-07-13 | 2015-12-02 | 哈尔滨工业大学 | Fault diagnosis method based on metric learning and time sequence during industrial process |
| CN106483929A (en) * | 2016-10-18 | 2017-03-08 | 重庆大学 | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means |
| CN107239789A (en) * | 2017-05-09 | 2017-10-10 | 浙江大学 | A kind of industrial Fault Classification of the unbalanced data based on k means |
| CN107368014A (en) * | 2017-08-24 | 2017-11-21 | 国网黑龙江省电力有限公司信息通信公司 | Machine room monitoring system and method |
-
2019
- 2019-07-05 CN CN201910605879.5A patent/CN110309885A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7512275B2 (en) * | 2003-10-21 | 2009-03-31 | Sharp Kabushiki Kaisha | Image collating apparatus, image collating method, image collating program and computer readable recording medium recording image collating program |
| CN101582195A (en) * | 2009-06-09 | 2009-11-18 | 深圳中兴力维技术有限公司 | Method for generating alarm in dynamic environment monitoring system |
| CN103218464A (en) * | 2013-03-23 | 2013-07-24 | 中国水利电力物资有限公司 | Wind turbine generator state-based wind turbine generator data storage method and system |
| CN105116872A (en) * | 2015-07-13 | 2015-12-02 | 哈尔滨工业大学 | Fault diagnosis method based on metric learning and time sequence during industrial process |
| CN106483929A (en) * | 2016-10-18 | 2017-03-08 | 重庆大学 | A kind of fault of numerical control machine tool cancellation scheme differentiates indicating means |
| CN107239789A (en) * | 2017-05-09 | 2017-10-10 | 浙江大学 | A kind of industrial Fault Classification of the unbalanced data based on k means |
| CN107368014A (en) * | 2017-08-24 | 2017-11-21 | 国网黑龙江省电力有限公司信息通信公司 | Machine room monitoring system and method |
Non-Patent Citations (2)
| Title |
|---|
| 李弼程等: "《网络舆情分析：理论、技术与应对策略》", 31 March 2015, 北京：国防工业出版社 * |
| 黄祥林等: "《图像检索原理与实践》", 30 June 2014, 北京：中国传媒大学出版社 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111209944A (en) * | 2019-12-31 | 2020-05-29 | 上海索广电子有限公司 | Self-adaptive image detection system and image detection method |
| CN111209944B (en) * | 2019-12-31 | 2023-09-01 | 上海索广映像有限公司 | Self-adaptive image detection system and image detection method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110287347A (en) | Using the method for big data detection electric power computer room failure | |
| CN105245185B (en) | A kind of area distribution formula photovoltaic fault diagnosis system and method for accessing power distribution network | |
| CN105915398B (en) | Rural power grid fault based rapid detection and power restoration system and concentrator detection method thereof | |
| CN103792420B (en) | Based on the anti-electricity-theft of load curve and supervision of power consumption method | |
| CN104218676B (en) | The intelligent warning system of power dispatching automation main website and method | |
| WO2020134361A1 (en) | State evaluation method for secondary equipment of substation, system, and equipment | |
| CN105656698A (en) | Intelligent monitoring structure and method for network application system | |
| CN108287294A (en) | Distribution network failure region Fast Identification Method based on power failure distribution transforming and topological analysis | |
| CN108923542A (en) | A kind of VR system for power scheduling | |
| CN110349048B (en) | Substation multidimensional data operation interactive control platform and fault handling method | |
| CN117078017A (en) | Intelligent decision analysis system for monitoring power grid equipment | |
| CN109709389A (en) | For electric instrument distributed mass real time data sampling alarm method and system | |
| CN113806420A (en) | Power grid data monitoring method and device | |
| CN111882110A (en) | Power information communication fault prediction system based on big data technology | |
| CN107658980A (en) | A kind of analysis method and system for being used to check power system monitor warning information | |
| CN116094174A (en) | Knowledge graph-based power grid operation and maintenance monitoring method, system, equipment and medium | |
| CN104834305B (en) | Distribution automation terminal remote measurement exception analysis system and method based on DMS systems | |
| CN115603459A (en) | Digital twin technology-based power distribution network key station monitoring method and system | |
| CN101995854A (en) | Monitoring method and system for ensuring safe operation of electric network | |
| CN110309885A (en) | Computer room state judging method based on big data | |
| CN107295553A (en) | A kind of base station maintenance method and device | |
| CN111969709B (en) | Intelligent alarm method and system based on power system spare power automatic switching analysis | |
| CN107102633B (en) | A kind of distribution terminal fault self-diagnosis method and system | |
| CN117394535A (en) | Digital twin system of AC/DC hybrid power distribution network | |
| CN108896100A (en) | A kind of capacitor group status monitoring subsystem |
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: 20191008 |