CN110033527A - A kind of electric operating security control intelligent robot and its implementation - Google Patents
A kind of electric operating security control intelligent robot and its implementation Download PDFInfo
- Publication number
- CN110033527A CN110033527A CN201910183167.9A CN201910183167A CN110033527A CN 110033527 A CN110033527 A CN 110033527A CN 201910183167 A CN201910183167 A CN 201910183167A CN 110033527 A CN110033527 A CN 110033527A
- Authority
- CN
- China
- Prior art keywords
- electric operating
- work order
- case
- regulations
- work
- 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
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C1/00—Registering, indicating or recording the time of events or elapsed time, e.g. time-recorders for work people
- G07C1/20—Checking timed patrols, e.g. of watchman
-
- G—PHYSICS
- G07—CHECKING-DEVICES
- G07C—TIME OR ATTENDANCE REGISTERS; REGISTERING OR INDICATING THE WORKING OF MACHINES; GENERATING RANDOM NUMBERS; VOTING OR LOTTERY APPARATUS; ARRANGEMENTS, SYSTEMS OR APPARATUS FOR CHECKING NOT PROVIDED FOR ELSEWHERE
- G07C3/00—Registering or indicating the condition or the working of machines or other apparatus, other than vehicles
- G07C3/08—Registering or indicating the production of the machine either with or without registering working or idle time
Abstract
The invention discloses a kind of electric operating security control intelligent robot and its implementation, through the invention, electric power enterprise accredits electric operating security control intelligent robot in operation field, security warning and report can be carried out because of act of violating regulations to the people that electric operating scene is found, the security control function at electric operating scene is realized, to solve the problems, such as that electric power enterprise security control ability and human and material resources are insufficient；Can recorde the complete operation process data in electric operating scene, convenient for people because caused by act of violating regulations accident event carry out ex-post analysis summary, be conducive to the quality and level that promote electric power enterprise security control.
Description
Technical field
The present invention relates to intelligent robot and its implementation, especially a kind of electric operating security control intelligent robot
And its implementation, belong to technical field.
Background technique
In power generation and the Supervision on Bio-safety of operation and maintenance, with the raising of power equipment reliability, people is not
The safety behavior principal risk source (violating the regulations) as safe and stable operation of power system.Electric power accident event can be divided into equipment thing
Story part, people are because of unprofessional accident event and contingent event, and in the accident event that these lead to personal injury, people is because violating the regulations
Accident event is the accident event that can be avoided or reduced by reinforcing security control.
Cause people because there are many factor violating the regulations, such as electric operating complexity factor, personnel itself factor, natural environment
Factor, weather condition factor, operation Work tool carry factor and operation field security control factor etc..It will be in these complexity
Factor in carry out the security control at electric operating scene, need high-caliber ability to supervise and enough manpower and material resources
It supports, and electric operating security control intelligent robot can make up electric power enterprise ability to supervise and human and material resources not
Foot.
Summary of the invention
The purpose of the present invention is to solve the defects of the above-mentioned prior art, provide a kind of electric operating peace of electric power enterprise
Full supervision intelligent robot and its implementation.
The purpose of the present invention can be reached by adopting the following technical scheme that:
A kind of electric operating security control intelligent robot, it is characterised in that: including electric operating work order violating the regulations, accident event and
Normal operation case library management module, electric operating are broken rules and regulations, accident event and normal operation pre-control analyze machine learning mould
Block, live electric operating work order information input module, electric operating personnel on site's ongoing operation attention test module, electric power are made
Industry site safety Work tool detection module, electric operating scene meteorological information acquisition module, electric operating field speech video figure
As information acquisition module, electric operating process safety alarm module, electric operating completion profiling module and electric operating safety
Supervise control module, the electric operating security control control module violating the regulations, accident event and just with electric operating work order respectively
Normal job history case database management module, electric operating be violating the regulations, accident event and normal operation pre-control analysis machine study module,
Live electric operating work order information input module, electric operating personnel on site's ongoing operation attention test module, electric operating
Site safety Work tool detection module, electric operating scene meteorological information acquisition module, electric operating field speech video image
Information acquisition module, electric operating process safety alarm module, electric operating completion profiling module sequence are connected, and realize to function
The operation of module controls.
As a preferred embodiment, the electric operating work order is violating the regulations, accident event and normal operation case library pipe
Reason module includes voice mode, handwriting mode, keyboard mode and data-interface mode, realizes that electric operating work order is violating the regulations, thing
The maintenance of story part and normal operation history case database.
As a preferred embodiment, the live electric operating work order information input module includes voice mode, hand-written side
Formula, keyboard mode and data-interface mode realize the acquisition of live electric operating work order information.
As a preferred embodiment, the electric operating site safety Work tool detection module detects skill using less radio-frequency
Art, the managing security tools and instruments carried to live electric operating personnel detect.
As a preferred embodiment, the electric operating scene meteorological information acquisition module uses weather monitoring technology, adopts
Collect operation field weather information.
As a preferred embodiment, the electric operating field speech video image acquisition module uses voice and video image
Technology, collection site voice and video image information.
A kind of implementation method of electric operating security control intelligent robot, which comprises the following steps:
Step 1: opening electric operating security control intelligent robot power switch, start electric operating security control control method
Program；
Step 2: starting electric operating security control intelligent robot initial work, comprising: electric operating work order is violating the regulations, accident
Event and the work of normal operation history case data storage, the setting work of electric operating security control control parameter and electric power are made
Industry is broken rules and regulations, work is arranged in accident event and the analysis machine learning model training of normal operation pre-control, if the work of this step is
It completes, then skips this step and directly execute step 3 work；
Step 3: according to electric operating, violating the regulations, accident event and normal operation pre-control analyze machine learning model training condition, instruction
Practice electric operating violating the regulations, accident event and normal operation pre-control analysis machine learning model to jump if training condition is unsatisfactory for
It crosses this step and directly executes step 4 work；
Step 4: starting live electric operating work order information and import program, complete the importing work of live electric operating work order information
Make；
Step 5: starting electric operating scene attention testing sequence prompts live electric operating personnel using natural language, leads to
It crosses portable mobile apparatus to fulfil assignment the test of live attention, and recycles all electric operating personnel in waiting scene and complete to pay attention to
Until when power test work；
Step 6: the time interval period of live first security control of starting electric operating；
Step 7: starting electric operating site safety Work tool detects program, completes live electric operating personnel work device safe to carry
The detection work of tool；
Step 8: starting electric operating scene meteorological information acquisition program completes live weather, temperature, humidity, wind-force and wind
It works to meteorological data collection；
Step 9: starting electric operating field speech video image acquisition program completes electric operating field speech video image letter
Cease collecting work；
Step 10: starting electric operating process safety warns program, uses operating personnel's attention of electric operating collection in worksite
Test information, managing security tools and instruments information, electric operating scene weather information and the voice and video figure that electric operating personnel carry
As information, power application operation is broken rules and regulations, accident event and normal operation pre-control analyze machine learning model, to live electric operating
Work order is broken rules and regulations, accident event and normal operation pre-control are analyzed, and pre-control analysis result is carried out site safety warning and report
It accuses；
Step 11: starting electric operating completion filing procedure, monitoring man-machine interactive interface input, if there is " operation completion filing "
Input information be then transferred to step 13, otherwise continue to execute step 12 work；
Step 12: waiting electric operating scene next security control time interval period, be transferred to step 7；
Step 13: prompt selection man-machine interactive interface, typing scene electric operating work order implementing result, and return according to operation completion
Shelves condition, violating the regulations, accident event and normal operation history case by live electric operating work order data deposit electric operating work order
In database；
Step 14: exiting electric operating security control control program, and close electric operating security control intelligent robot power supply.
Wherein, the step 2 of the method further includes the following steps:
Step 21: using voice mode or handwriting mode or keyboard mode or data-interface mode, importing electric operating work order and disobey
Chapter, accident event and normal operation history case data complete electric operating work order violating the regulations, accident event and normal operation history
Case data is put in storage work；
Step 22: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating and supervise safely
The parameter setting work of electric operating scene weather monitoring area name in pipe；
Step 23: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating process peace
Work is arranged by the gap periods threshold parameter of unit timing of minute in full supervision；
Step 24: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing that electric operating is violating the regulations, thing
Story part and normal operation pre-control analysis machine learning model training condition are that history case data increased number is more than or equal to certain
A specified threshold parameter setting work；
Step 25: using voice mode or handwriting mode or keyboard mode or data-interface mode, complete electric operating it is violating the regulations and
Accident event information warning report the office phone number of safety regulator contact person, cell phone number, wechat number with
And e-mail contact method parameter setting work.
Wherein, the step 3 of the method further includes the following steps:
Step 31: defining the feature space of electric operating work order violating the regulations, accident event and normal operation history case data space D
T and its valued space X；
Step 32: in feature space T calculate feature valued space X association operate against regulations work order, accident event operation work order with
And the related information matrix table 1 of normal operation work order:
Step 33: the comentropy for defining case work order association in the D of case space violating the regulations, accident event and normal work order calculates public affairs
Formula:
Define 1: case work order diAssociation comentropy calculation formula violating the regulations
H1(di)=∑j=1…M Uj*log(1/Pj,k), Pj,k = Aj,k/E jFor case work order diThe k-th value of feature j exist
The corresponding work order ratio violating the regulations of the k-th value of 1 feature j of related information matrix table；
Define 2: case work order diIt is associated with the comentropy calculation formula of accident event
H2(di)=∑j=1…M Uj*log(1/Qj,k), Qj,k = Bj,k/FjFor case work order diThe k-th value of feature j exist
The corresponding accident event work order ratio of k-th value of 1 feature j of related information matrix table；
Define 3: case work order diIt is associated with the comentropy calculation formula of normal work order
H3(di)=∑j=1…M Uj*log(1/Rj,k), Rj,k = Cj,k/GjFor case work order diThe k-th value of feature j exist
The corresponding normal work order ratio of the k-th value of 1 feature j of related information matrix table；
In above-mentioned definition, M is characterized the dimension of space T, log (1/Pj,k)、log(1/Qj,k) and log (1/Rj,k) be with 2 be bottom
Logarithmic function, Uj=Kj/ TN, KjFor the value number of jth dimensional feature, TN be characterized the value sum TN of space T feature=
∑j=1…M jk；
Step 34: calculate the comentropy matrix table 2 of case work order in the D of case space:
Step 35: defining any two case d in the D of case spaceiWith djRange formula are as follows: | didj|=；
Wherein, defining case space D and carrying out k-means clustering learning and the ratio of cross validation is u:v, and u is to carry out k- in D
The case work order ratio of means clustering learning, v are the case work order ratio that cross validation is carried out in D.It is random in the ratio of u:v
Ground extracts case work order from D and constitutes the case space D1 of clustering learning and constitute the case space D2 of cross validation, accordingly
Comentropy spatial table 2 is split as to the comentropy spatial table 2-1 and cross validation case space D2 of clustering learning case space D1
Comentropy spatial table 2-2:
Wherein nu:nv=u:v；
Step 36: the k value (k > 1) of selected k-means clustering learning；
Step 37: using k-means clustering algorithm to case work order d in comentropy spatial table 2-1iBy distance | didj| it carries out
Cluster, k initial cluster center randomly select from comentropy spatial table 2-1, cluster j (j=1,2 ... cluster k) when iteration
The calculation formula of central point are as follows:
H1(j)= ∑i=1…N1j H1(di)/N1j,
H2(j)= ∑i=1…N2j H2(di)/N2j,
H3(j)= ∑i=1…N3j H3(di)/N3j,
Wherein, N1jFor the number for case work order of breaking rules and regulations in cluster j, H1 (di) it is work order d in cluster jiViolation information entropy, N2j
For the number of accident event case work order in cluster j, H2 (di) it is work order d in cluster jiAccident event comentropy, N3jIt is poly-
The number of normal case work order, H3 (d in class ji) it is work order d in cluster j iNormal work order information entropy；(H1 (j), H2 (j), H3
(j)) cluster centre to cluster j when cluster iteration；Using k-means clustering algorithm to case in comentropy spatial table 2-1
Work order diBy distance | didj| it is clustered, obtains k clustering learning model Ms as shown in table 31 ,M2,…, Mk:
Wherein, N1, N2, N3 are respectively that clustering learning case space D1 breaks rules and regulations case work order, accident event case work order and just
The quantity of normal case work order；
Step 38: using the comentropy matrix table 2-2 of cross validation case space D2
To k clustering learning model M={ M of table 31 ,M2,…,MkCarry out cross validation, cross validation algorithm are as follows:
1) setting b=0(b is to verify the practical implementing result of case work order in D2 and gather
Class learning model M={ M1 ,M2,…,MkExpected from the quantity that is consistent of implementing result)；
2) i=1 is set；
3) i-th in D2 of verifying case work order dd is takeni；
4) dd is calculated separately outiWith clustering learning model M={ M1 ,M2,…,Mk}
K distance value VV of K cluster centre1, VV2。。。VVK, case work order dd will be verifiediIt is included into the smallest distance value VVjIt is right
The clustering learning model M answeredjIn；
5) clustering learning model M is chosenjIn break rules and regulations case ratio, accident event ratio
And the maximum in normal case ratio three is as clustering learning model MjTo verifying case work order ddiExpected knot
Fruit (case, accident event case or normal case violating the regulations), if this expected results and cross validation case ddiActually hold
Row result is consistent then b=b+1, otherwise performs the next step；
6)i=i+1；
7) turn step 3 if i≤nv) otherwise perform the next step；
8) cross validation algorithm terminates.
Step 39: if cross validation results b/nv reaches clustering learning model desired effect or meets clustering learning mould
Type the number of iterations termination condition then goes to step 310, otherwise goes to step the clustering learning model training of a 36 carry out new rounds；
Step 310: step 3 algorithm of the method terminates.
The present invention have compared with the existing technology it is following the utility model has the advantages that
1, through the invention, electric power enterprise accredits electric operating security control intelligent robot in operation field, can be to electric power
The people of operation field discovery realizes the security control function at electric operating scene because act of violating regulations carries out security warning and report,
To solve the problems, such as that electric power enterprise security control ability and human and material resources are insufficient；
2, by the invention it is possible to the complete operation process data of record electricity operation field, convenient for people because act of violating regulations is made
At accident event carry out ex-post analysis summary, be conducive to promoted electric power enterprise security control quality and level.
Detailed description of the invention
Fig. 1 is the structure chart of robot of the invention.
Fig. 2 is the flow chart that robot of the invention works.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
A kind of electric operating security control intelligent robot and its implementation are further elaborated.It should be appreciated that this place
The specific embodiment of description is only used to explain the present invention, is not intended to limit the present invention.
Embodiment 1:
Fig. 1 is a kind of structure chart of electric operating security control intelligent robot provided by the invention, a kind of electricity of the present embodiment
Masterpiece industry security control intelligent robot, it is characterised in that: violating the regulations, accident event and normal operation are gone through including electric operating work order
History case database management module, electric operating are broken rules and regulations, accident event and normal operation pre-control analyze machine study module, live electric power
Operation work order information input module, electric operating personnel on site's ongoing operation attention test module, electric operating site safety
Work tool detection module, electric operating scene meteorological information acquisition module, the acquisition of electric operating field speech video image information
Module, electric operating process safety alarm module, electric operating completion profiling module and electric operating security control control mould
Block, the electric operating security control control module violating the regulations, accident event and normal operation history with electric operating work order respectively
Case database management module, electric operating are broken rules and regulations, accident event and normal operation pre-control analysis machine study module, live electric power are made
Industry work order information input module, electric operating personnel on site's ongoing operation attention test module, electric operating site safety work
Utensil detection module, electric operating scene meteorological information acquisition module, electric operating field speech video image information acquire mould
Block, electric operating process safety alarm module, electric operating completion profiling module sequence are connected, and realize and disobey to electric operating work order
Chapter, accident event and the maintenance of normal operation history case database；Realize violating the regulations, accident event and normal operation to electric operating
Pre-control analyzes machine learning model training；It realizes and live electric operating work order information and live electric operating personnel is persistently grasped
Make the acquisition of attention test information；It realizes and information, electric operating scene meteorology is carried to live electric operating managing security tools and instruments
The acquisition of information and electric operating field speech video image information；Violating the regulations, accident event and normal work using electric operating
Industry pre-control analysis machine learning model breaks rules and regulations to live electric operating work order, accident event and normal operation pre-control are analyzed,
And field speech security warning and report are carried out to pre-control analysis result in operation process；It realizes to the complete wage of electric operating work order
Expect electric operating work order is violating the regulations, the filing processing in accident event and normal operation case library；Realize that electric operating is existing
Field weather monitoring regional parameters, electric operating process safety supervise time interval cycle parameter, electric operating is broken rules and regulations, accident event
And normal operation pre-control analysis machine learning model training condition parameter, electric operating are violating the regulations and accident event warning report is safe
The setting of supervision department's contact method parameter, electric operating work order completion archive condition parameter, and realize the fortune of each functional module
Row control.
Each module concrete function and implementation are described as follows:
Electric operating work order is violating the regulations, accident event and normal operation case library management module: including voice mode, hand-written side
Formula, keyboard mode and data-interface mode realize electric operating work order violating the regulations, accident event and normal operation history case number of cases
According to the maintenance in library.
Electric operating is broken rules and regulations, accident event and normal operation pre-control analyze machine study module: power application operation work order
Violating the regulations, accident event and normal operation history case data realize electric operating violating the regulations, accident event and normal operation pre-control point
Analyse the training of machine learning model.
Live electric operating work order information input module: it is connect including voice mode, handwriting mode, keyboard mode and data
Mouth mode realizes the acquisition of live electric operating work order information.
Electric operating personnel on site's ongoing operation attention test module: applying ongoing operation attention test blueprint, right
Live electric operating personnel carry out the test of ongoing operation attention.
Electric operating site safety Work tool detection module: less radio-frequency detection technique is applied, to live electric operating people
The managing security tools and instruments that member carries is detected.
Electric operating scene meteorological information acquisition module: applicating atmosphere monitoring technology, Collecting operation scene weather information.
Electric operating field speech video image acquisition module: voice and video image technique, collection site voice view are applied
Frequency image information.
Electric operating process safety alarm module: the operating personnel scene ongoing operation according to electric operating collection in worksite is infused
Power of anticipating test information, the managing security tools and instruments information of live electric operating personnel carrying, electric operating scene weather information, electric power are made
Industry field speech and video image information and live electric operating work order information, power application operation is violating the regulations, accident event and
Normal operation pre-control analysis machine learning model breaks rules and regulations to live electric operating work order, accident event and normal operation it is pre-
Control analysis, and security warning and report are carried out to pre-control analysis result.
Electric operating completion profiling module: violating the regulations, accident event and the filing of normal operation history case according to electric operating
Condition carries out filing processing to live electric operating work order.
Electric operating security control control method module: applying natural language processing technique and input/output interface technology,
Realize that electric operating scene weather monitoring regional parameters, electric operating process safety supervision time interval cycle parameter, electric power are made
Industry is violating the regulations, accident event and normal operation pre-control analysis machine learning model training condition parameter, electric operating be violating the regulations and accident
The setting and reality of event warning report safety regulator's contact method parameter, electric operating work order completion archive condition parameter
The human-computer interaction of existing electric operating process, and realize the operation control of each functional module.
Embodiment 2:
As shown in Figure 1, a kind of implementation method of electric operating security control intelligent robot of the present embodiment, which is characterized in that
The following steps are included:
Step 1: opening electric operating security control intelligent robot power switch, start electric operating security control control method
Program；
Step 2: starting electric operating security control intelligent robot initial work, comprising: electric operating work order is violating the regulations, accident
Event and the work of normal operation history case data storage, the setting work of electric operating security control control parameter and electric power are made
Industry is broken rules and regulations, work is arranged in accident event and the analysis machine learning model training of normal operation pre-control, if the work of this step is
It completes, then skips this step and directly execute step 3 work；
Step 3: according to electric operating, violating the regulations, accident event and normal operation pre-control analyze machine learning model training condition, instruction
Practice electric operating violating the regulations, accident event and normal operation pre-control analysis machine learning model to jump if training condition is unsatisfactory for
It crosses this step and directly executes step 4 work；
Step 4: starting live electric operating work order information and import program, complete the importing work of live electric operating work order information
Make；
Step 5: starting electric operating scene attention testing sequence prompts live electric operating personnel using natural language, leads to
It crosses portable mobile apparatus to fulfil assignment the test of live attention, and recycles all electric operating personnel in waiting scene and complete to pay attention to
Until when power test work；
Step 6: the time interval period of live first security control of starting electric operating；
Step 7: starting electric operating site safety Work tool detects program, completes live electric operating personnel work device safe to carry
The detection work of tool；
Step 8: starting electric operating scene meteorological information acquisition program completes live weather, temperature, humidity, wind-force and wind
It works to meteorological data collection；
Step 9: starting electric operating field speech video image acquisition program completes electric operating field speech video image letter
Cease collecting work；
Step 10: starting electric operating process safety warns program, uses operating personnel's attention of electric operating collection in worksite
Test information, managing security tools and instruments information, electric operating scene weather information and the voice and video figure that electric operating personnel carry
As information, power application operation is broken rules and regulations, accident event and normal operation pre-control analyze machine learning model, to live electric operating
Work order is broken rules and regulations, accident event and normal operation pre-control are analyzed, and pre-control analysis result is carried out site safety warning and report
It accuses；
Step 11: starting electric operating completion filing procedure, monitoring man-machine interactive interface input, if there is " operation completion filing "
Input information be then transferred to step 13, otherwise continue to execute step 12 work；
Step 12: waiting electric operating scene next security control time interval period, be transferred to step 7；
Step 13: prompt selection man-machine interactive interface, typing scene electric operating work order implementing result, and return according to operation completion
Shelves condition, violating the regulations, accident event and normal operation history case by live electric operating work order data deposit electric operating work order
In database；
Step 14: exiting electric operating security control control program, and close electric operating security control intelligent robot power supply.
Wherein, the step 2 of the method further includes the following steps:
Step 21: using voice mode or handwriting mode or keyboard mode or data-interface mode, importing electric operating work order and disobey
Chapter, accident event and normal operation history case data complete electric operating work order violating the regulations, accident event and normal operation history
Case data is put in storage work；
Step 22: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating and supervise safely
The parameter setting work of electric operating scene weather monitoring area name in pipe；
Step 23: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating process peace
Work is arranged by the gap periods threshold parameter of unit timing of minute in full supervision；
Step 24: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing that electric operating is violating the regulations, thing
Story part and normal operation pre-control analysis machine learning model training condition are that history case data increased number is more than or equal to certain
A specified threshold parameter setting work；
Step 25: using voice mode or handwriting mode or keyboard mode or data-interface mode, complete electric operating it is violating the regulations and
Accident event information warning report the office phone number of safety regulator contact person, cell phone number, wechat number with
And e-mail contact method parameter setting work.
Wherein, the step 3 of the method further includes the following steps:
Step 31: defining the feature space of electric operating work order violating the regulations, accident event and normal operation history case data space D
T and its valued space X；
Step 32: in feature space T calculate feature valued space X association operate against regulations work order, accident event operation work order with
And the related information matrix table 1 of normal operation work order:
Step 33: the comentropy for defining case work order association in the D of case space violating the regulations, accident event and normal work order calculates public affairs
Formula:
Define 1: case work order diAssociation comentropy calculation formula violating the regulations
H1(di)=∑j=1…M Uj*log(1/Pj,k), Pj,k = Aj,k/EjFor case work order diThe k-th value of feature j exist
The corresponding work order ratio violating the regulations of the k-th value of 1 feature j of related information matrix table；
Define 2: case work order diIt is associated with the comentropy calculation formula of accident event
H2(di)=∑j=1…M Uj*log(1/Qj,k), Qj,k = Bj,k/FjFor case work order diThe k-th value of feature j exist
The corresponding accident event work order ratio of k-th value of 1 feature j of related information matrix table；
Define 3: case work order diIt is associated with the comentropy calculation formula of normal work order
H3(di)=∑j=1…M Uj*log(1/Rj,k), Rj,k = Cj,k/GjFor case work order diThe k-th value of feature j exist
The corresponding normal work order ratio of the k-th value of 1 feature j of related information matrix table；
In above-mentioned definition, M is characterized the dimension of space T, log (1/Pj,k)、log(1/Qj,k) and log (1/Rj,k) be with 2 be bottom
Logarithmic function, Uj=Kj/ TN, KjFor the value number of jth dimensional feature, TN be characterized the value sum TN of space T feature=
∑j=1…M jk；
Step 34: calculate the comentropy matrix table 2 of case work order in the D of case space:
Step 35: defining any two case d in the D of case spaceiWith djRange formula are as follows: | didj|=；
Wherein, defining case space D and carrying out k-means clustering learning and the ratio of cross validation is u:v, and u is to carry out k- in D
The case work order ratio of means clustering learning, v are the case work order ratio that cross validation is carried out in D.It is random in the ratio of u:v
Ground extracts case work order from D and constitutes the case space D1 of clustering learning and constitute the case space D2 of cross validation, accordingly
Comentropy spatial table 2 is split as to the comentropy spatial table 2-1 and cross validation case space D2 of clustering learning case space D1
Comentropy spatial table 2-2:
Wherein nu:nv=u:v；
Step 36: the k value (k > 1) of selected k-means clustering learning；
Step 37: using k-means clustering algorithm to case work order d in comentropy spatial table 2-1iBy distance | didj| it carries out
Cluster, k initial cluster center randomly select from comentropy spatial table 2-1, cluster j (j=1,2 ... cluster k) when iteration
The calculation formula of central point are as follows:
H1(j)= ∑i=1…N1j H1(di)/N1j,
H2(j)= ∑i=1…N2j H2(di)/N2j,
H3(j)= ∑i=1…N3j H3(di)/N3j,
Wherein, N1jFor the number for case work order of breaking rules and regulations in cluster j, H1 (di) it is work order d in cluster jiViolation information entropy, N2j
For the number of accident event case work order in cluster j, H2 (di) it is work order d in cluster jiAccident event comentropy, N3jIt is poly-
The number of normal case work order, H3 (d in class ji) it is work order d in cluster jiNormal work order information entropy；(H1 (j), H2 (j), H3
(j)) cluster centre to cluster j when cluster iteration；Using k-means clustering algorithm to case in comentropy spatial table 2-1
Work order diBy distance | didj| it is clustered, obtains k clustering learning model Ms as shown in table 31 ,M2,…, Mk:
Wherein, N1, N2, N3 are respectively that clustering learning case space D1 breaks rules and regulations case work order, accident event case work order and just
The quantity of normal case work order；
Step 38: using the comentropy matrix table 2-2 of cross validation case space D2
To k clustering learning model M={ M of table 31 ,M2,…,MkCarry out cross validation, cross validation algorithm are as follows:
1) setting b=0(b is to verify the practical implementing result of case work order in D2 and gather
Class learning model M={ M1 ,M2,…,MkExpected from the quantity that is consistent of implementing result)；
2) i=1 is set；
3) i-th in D2 of verifying case work order dd is takeni；
4) dd is calculated separately outiWith clustering learning model M={ M1 ,M2,…,Mk}
K distance value VV of K cluster centre1, VV2。。。VVK, case work order dd will be verifiediIt is included into the smallest distance value VVjIt is right
The clustering learning model M answeredjIn；
5) clustering learning model M is chosenjIn break rules and regulations case ratio, accident event ratio
And the maximum in normal case ratio three is as clustering learning model MjTo verifying case work order ddiExpected knot
Fruit (case, accident event case or normal case violating the regulations), if this expected results and cross validation case ddiActually hold
Row result is consistent then b=b+1, otherwise performs the next step；
6)i=i+1；
7) turn step 3 if i≤nv) otherwise perform the next step；
8) cross validation algorithm terminates.
Step 39: if cross validation results b/nv reaches clustering learning model desired effect or meets clustering learning mould
Type the number of iterations termination condition then goes to step 310, otherwise goes to step the clustering learning model training of a 36 carry out new rounds；
Step 310: step 3 algorithm of the method terminates.
The present invention is using private server computer as information process unit；Case data built in private server computer
Library is using solid state hard disk as storage medium；Security control to electric operating is completed by multilayer architecture software and hardware function
, software and hardware function has upgrading ability.Therefore, a kind of electric operating security control intelligent robot of the invention and its realization
Method has the flexibility to update.
The above, only the invention patent preferred embodiment, but the scope of protection of the patent of the present invention is not limited to
This, anyone skilled in the art is in the range disclosed in the invention patent, according to the present invention the skill of patent
Art scheme and its patent of invention design are subject to equivalent substitution or change, belong to the scope of protection of the patent of the present invention.
Claims (9)
1. a kind of electric operating security control intelligent robot, it is characterised in that: the violating the regulations, accident event including electric operating work order
And normal operation case library management module, electric operating are violating the regulations, accident event and normal operation pre-control analyze machine learning
Module, live electric operating work order information input module, electric operating personnel on site's ongoing operation attention test module, electric power
Operation field managing security tools and instruments detection module, electric operating scene meteorological information acquisition module, electric operating field speech video
Image information collecting module, electric operating process safety alarm module, electric operating completion profiling module and electric operating peace
Full supervision control module, the electric operating security control control module respectively with electric operating work order violating the regulations, accident event and
Normal operation case library management module, electric operating are broken rules and regulations, accident event and normal operation pre-control analyze machine learning mould
Block, live electric operating work order information input module, electric operating personnel on site's ongoing operation attention test module, electric power are made
Industry site safety Work tool detection module, electric operating scene meteorological information acquisition module, electric operating field speech video figure
As information acquisition module, electric operating process safety alarm module, electric operating complete, profiling module sequence is connected, and realization is to function
The operation control of energy module.
2. a kind of electric operating security control intelligent robot according to claim 1, it is characterised in that: the electric power is made
Industry work order is broken rules and regulations, accident event and normal operation case library management module include voice mode, handwriting mode, keyboard mode
And data-interface mode, realize that electric operating work order is violating the regulations, maintenance of accident event and normal operation history case database.
3. a kind of electric operating security control intelligent robot according to claim 1, it is characterised in that: the scene electricity
Masterpiece industry work order information input module includes voice mode, handwriting mode, keyboard mode and data-interface mode, realizes scene
The acquisition of electric operating work order information.
4. a kind of electric operating security control intelligent robot according to claim 1, it is characterised in that: the electric power is made
Industry site safety Work tool detection module uses less radio-frequency detection technique, the safe work device carried to live electric operating personnel
Tool is detected.
5. a kind of electric operating security control intelligent robot according to claim 1, it is characterised in that: the electric power is made
Industry scene meteorological information acquisition module uses weather monitoring technology, Collecting operation scene weather information.
6. a kind of electric operating security control intelligent robot according to claim 1, it is characterised in that: the electric power is made
Industry field speech video image acquisition module uses voice and video image technique, collection site voice and video image information.
7. a kind of implementation method of electric operating security control intelligent robot, which comprises the following steps:
Step 1: opening electric operating security control intelligent robot power switch, start electric operating security control control method
Program；
Step 2: starting electric operating security control intelligent robot initial work, comprising: electric operating work order is violating the regulations, accident
Event and the work of normal operation history case data storage, the setting work of electric operating security control control parameter and electric power are made
Industry is broken rules and regulations, work is arranged in accident event and the analysis machine learning model training of normal operation pre-control, if the work of this step is
It completes, then skips this step and directly execute step 3 work；
Step 3: according to electric operating, violating the regulations, accident event and normal operation pre-control analyze machine learning model training condition, instruction
Practice electric operating violating the regulations, accident event and normal operation pre-control analysis machine learning model to jump if training condition is unsatisfactory for
It crosses this step and directly executes step 4 work；
Step 4: starting live electric operating work order information and import program, complete the importing work of live electric operating work order information
Make；
Step 5: starting electric operating scene attention testing sequence prompts live electric operating personnel using natural language, leads to
It crosses portable mobile apparatus to fulfil assignment the test of live attention, and recycles all electric operating personnel in waiting scene and complete to pay attention to
Until when power test work；
Step 6: the time interval period of live first security control of starting electric operating；
Step 7: starting electric operating site safety Work tool detects program, completes live electric operating personnel work device safe to carry
The detection work of tool；
Step 8: starting electric operating scene meteorological information acquisition program completes live weather, temperature, humidity, wind-force and wind
It works to meteorological data collection；
Step 9: starting electric operating field speech video image acquisition program completes electric operating field speech video image letter
Cease collecting work；
Step 10: starting electric operating process safety warns program, uses operating personnel's attention of electric operating collection in worksite
Test information, managing security tools and instruments information, electric operating scene weather information and the voice and video figure that electric operating personnel carry
As information, power application operation is broken rules and regulations, accident event and normal operation pre-control analyze machine learning model, to live electric operating
Work order is broken rules and regulations, accident event and normal operation pre-control are analyzed, and pre-control analysis result is carried out site safety warning and report
It accuses；
Step 11: starting electric operating completion filing procedure, monitoring man-machine interactive interface input, if there is " operation completion filing "
Input information be then transferred to step 13, otherwise continue to execute step 12 work；
Step 12: waiting electric operating scene next security control time interval period, be transferred to step 7；
Step 13: prompt selection man-machine interactive interface, typing scene electric operating work order implementing result, and return according to operation completion
Shelves condition, violating the regulations, accident event and normal operation history case by live electric operating work order data deposit electric operating work order
In database；
Step 14: exiting electric operating security control control program, and close electric operating security control intelligent robot power supply.
8. a kind of implementation method of electric operating security control intelligent robot according to claim 7, it is characterised in that:
The step 2 of the method further includes the following steps:
Step 21: using voice mode or handwriting mode or keyboard mode or data-interface mode, importing electric operating work order and disobey
Chapter, accident event and normal operation history case data complete electric operating work order violating the regulations, accident event and normal operation history
Case data is put in storage work；
Step 22: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating and supervise safely
The parameter setting work of electric operating scene weather monitoring area name in pipe；
Step 23: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing electric operating process peace
Work is arranged by the gap periods threshold parameter of unit timing of minute in full supervision；
Step 24: using voice mode or handwriting mode or keyboard mode or data-interface mode, completing that electric operating is violating the regulations, thing
Story part and normal operation pre-control analysis machine learning model training condition are that history case data increased number is more than or equal to certain
A specified threshold parameter setting work；
Step 25: using voice mode or handwriting mode or keyboard mode or data-interface mode, complete electric operating it is violating the regulations and
Accident event information warning report the office phone number of safety regulator contact person, cell phone number, wechat number with
And e-mail contact method parameter setting work.
9. a kind of implementation method of electric operating security control intelligent robot according to claim 7, it is characterised in that:
The step 3 of the method further includes the following steps:
Step 31: defining the feature space of electric operating work order violating the regulations, accident event and normal operation history case data space D
T and its valued space X；
Step 32: in feature space T calculate feature valued space X association operate against regulations work order, accident event operation work order with
And the related information matrix table 1 of normal operation work order:
Step 33: the comentropy for defining case work order association in the D of case space violating the regulations, accident event and normal work order calculates public affairs
Formula:
Define 1: case work order diAssociation comentropy calculation formula violating the regulations
H1(di)=∑j=1…M Uj*log(1/Pj,k), Pj,k = Aj,k/EjFor case work order diThe k-th value of feature j closing
Join the corresponding work order ratio violating the regulations of k-th value of 1 feature j of information matrix table；
Define 2: case work order diIt is associated with the comentropy calculation formula of accident event
H2(di)=∑j=1…M Uj*log(1/Qj,k), Qj,k = Bj,k/FjFor case work order diThe k-th value of feature j closing
Join the corresponding accident event work order ratio of k-th value of 1 feature j of information matrix table；
Define 3: case work order diIt is associated with the comentropy calculation formula of normal work order
H3(di)=∑j=1…M Uj*log(1/Rj,k), Rj,k = Cj,k/GjFor case work order diThe k-th value of feature j closing
Join the corresponding normal work order ratio of k-th value of 1 feature j of information matrix table；
In above-mentioned definition, M is characterized the dimension of space T, log (1/Pj,k)、log(1/Qj,k) and log (1/Rj,k) be with 2 be bottom
Logarithmic function, Uj=Kj/ TN, KjFor the value number of jth dimensional feature, TN be characterized the value sum TN of space T feature=
∑j=1…M jk；
Step 34: calculate the comentropy matrix table 2 of case work order in the D of case space:
Step 35: defining any two case d in the D of case spaceiWith djRange formula are as follows: | didj|= ；
Wherein, defining case space D and carrying out k-means clustering learning and the ratio of cross validation is u:v, and u is to carry out k- in D
The case work order ratio of means clustering learning, v are the case work order ratio that cross validation is carried out in D.It is random in the ratio of u:v
Ground extracts case work order from D and constitutes the case space D1 of clustering learning and constitute the case space D2 of cross validation, accordingly
Comentropy spatial table 2 is split as to the comentropy spatial table 2-1 and cross validation case space D2 of clustering learning case space D1
Comentropy spatial table 2-2:
Wherein nu:nv=u:v；
Step 36: the k value (k > 1) of selected k-means clustering learning；
Step 37: using k-means clustering algorithm to case work order d in comentropy spatial table 2-1iBy distance | didj| it carries out
Cluster, k initial cluster center randomly select from comentropy spatial table 2-1, cluster j (j=1,2 ... cluster k) when iteration
The calculation formula of central point are as follows:
H1(j)= ∑i=1…N1j H1(di)/N1j,
H2(j)= ∑i=1…N2j H2(di)/N2j,
H3(j)= ∑i=1…N3j H3(di)/N3j,
Wherein, N1jFor the number for case work order of breaking rules and regulations in cluster j, H1 (di) it is work order d in cluster j iViolation information entropy, N2j
For the number of accident event case work order in cluster j, H2 (di) it is work order d in cluster j iAccident event comentropy, N3jIt is poly-
The number of normal case work order, H3 (d in class ji) it is work order d in cluster j iNormal work order information entropy；(H1 (j), H2 (j), H3
(j)) cluster centre to cluster j when cluster iteration；Using k-means clustering algorithm to case in comentropy spatial table 2-1
Work order diBy distance | didj| it is clustered, obtains k clustering learning model Ms as shown in table 31 ,M2,…, Mk:
Wherein, N1, N2, N3 are respectively that clustering learning case space D1 breaks rules and regulations case work order, accident event case work order and just
The quantity of normal case work order；
Step 38: using the comentropy matrix table 2-2 of cross validation case space D2
To k clustering learning model M={ M of table 31 ,M2,…,MkCarry out cross validation, cross validation algorithm are as follows:
1) setting b=0(b is to verify the practical implementing result of case work order in D2 and gather
Class learning model M={ M1 ,M2,…,MkExpected from the quantity that is consistent of implementing result)；
2) i=1 is set；
3) i-th in D2 of verifying case work order dd is takeni；
4) dd is calculated separately outiWith clustering learning model M={ M1,M2,…,Mk}
K distance value VV of K cluster centre1, VV2。。。VVK, case work order dd will be verifiediIt is included into the smallest distance value VVjIt is right
The clustering learning model M answeredjIn；
5) clustering learning model M is chosenjIn break rules and regulations case ratio, accident event ratio
And the maximum in normal case ratio three is as clustering learning model MjTo verifying case work order ddiExpected knot
Fruit (case, accident event case or normal case violating the regulations), if this expected results and cross validation case ddiActually hold
Row result is consistent then b=b+1, otherwise performs the next step；
6)i=i+1；
7) turn step 3 if i≤nv) otherwise perform the next step；
8) cross validation algorithm terminates.
Step 39: changing if cross validation results b/nv reaches clustering learning model desired effect or meets clustering learning model
Generation number termination condition then goes to step 310, otherwise goes to step the clustering learning model training of a 36 carry out new rounds；
Step 310: step 3 algorithm of the method terminates.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910183167.9A CN110033527A (en) | 2019-03-12 | 2019-03-12 | A kind of electric operating security control intelligent robot and its implementation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910183167.9A CN110033527A (en) | 2019-03-12 | 2019-03-12 | A kind of electric operating security control intelligent robot and its implementation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110033527A true CN110033527A (en) | 2019-07-19 |
Family
ID=67235834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910183167.9A Pending CN110033527A (en) | 2019-03-12 | 2019-03-12 | A kind of electric operating security control intelligent robot and its implementation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110033527A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110865576A (en) * | 2019-11-20 | 2020-03-06 | 孙宗民 | Electric power construction operation safety management robot system |
| CN115115474A (en) * | 2022-08-29 | 2022-09-27 | 广东电网有限责任公司佛山供电局 | Electric power operation violation data analysis method and system |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2382555A1 (en) * | 2008-12-23 | 2011-11-02 | Andrew Wong | System, method and computer program for pattern based intelligent control, monitoring and automation |
| CN107256415A (en) * | 2017-08-04 | 2017-10-17 | 国网北京经济技术研究院 | A kind of computational methods and computing system of power system operation mode scene |
| CN108833831A (en) * | 2018-06-15 | 2018-11-16 | 陈在新 | A kind of power construction intelligent safety monitor system |
| CN109002897A (en) * | 2018-08-29 | 2018-12-14 | 国网河南省电力公司检修公司 | Field operation in transformer substation security management and control system and method based on robot technology |
| CN109344769A (en) * | 2018-09-29 | 2019-02-15 | 中国资源卫星应用中心 | A kind of photovoltaic plant detection method and system based on remote sensing image |
-
2019
- 2019-03-12 CN CN201910183167.9A patent/CN110033527A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2382555A1 (en) * | 2008-12-23 | 2011-11-02 | Andrew Wong | System, method and computer program for pattern based intelligent control, monitoring and automation |
| CN107256415A (en) * | 2017-08-04 | 2017-10-17 | 国网北京经济技术研究院 | A kind of computational methods and computing system of power system operation mode scene |
| CN108833831A (en) * | 2018-06-15 | 2018-11-16 | 陈在新 | A kind of power construction intelligent safety monitor system |
| CN109002897A (en) * | 2018-08-29 | 2018-12-14 | 国网河南省电力公司检修公司 | Field operation in transformer substation security management and control system and method based on robot technology |
| CN109344769A (en) * | 2018-09-29 | 2019-02-15 | 中国资源卫星应用中心 | A kind of photovoltaic plant detection method and system based on remote sensing image |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110865576A (en) * | 2019-11-20 | 2020-03-06 | 孙宗民 | Electric power construction operation safety management robot system |
| CN115115474A (en) * | 2022-08-29 | 2022-09-27 | 广东电网有限责任公司佛山供电局 | Electric power operation violation data analysis method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11551103B2 (en) | Data-driven activity prediction | |
| EP2069993A2 (en) | Security system and method for detecting intrusion in a computerized system | |
| CN109214448B (en) | Non-performance person training method, system, terminal and computer readable storage medium | |
| CN106548788A (en) | A kind of intelligent emotion determines method and system | |
| CN111614491A (en) | Power monitoring system oriented safety situation assessment index selection method and system | |
| CN110047512A (en) | A kind of ambient sound classification method, system and relevant apparatus | |
| CN109800995A (en) | A kind of grid equipment fault recognition method and system | |
| CN110033527A (en) | A kind of electric operating security control intelligent robot and its implementation | |
| WO2014173270A1 (en) | Human-machine interface detection method and system | |
| CN111160802A (en) | Method and device for evaluating preset scheme | |
| CN110533094A (en) | A kind of evaluation method and system for driver | |
| CN113744890A (en) | Reworking and production-resuming analysis method, system and storage medium | |
| CN106779404A (en) | A kind of pre- alarm method of flow and device | |
| CN113138894B (en) | Experimental equipment monitoring method based on power parameter monitoring and screen information identification | |
| CN113807462B (en) | Network equipment fault cause positioning method and system based on AI | |
| CN114201328A (en) | Fault processing method and device based on artificial intelligence, electronic equipment and medium | |
| CN113887799A (en) | Artificial intelligent alarm method for safety production of hydraulic power plant | |
| Bogen et al. | Evaluating a data clustering approach for life-cycle facility control. | |
| CN116701639B (en) | Text analysis-based double-carbon knowledge graph data analysis method and system | |
| CN116955592B (en) | Data processing method and system based on visual reasoning result | |
| CN116582339B (en) | Intelligent building network security monitoring method and monitoring system | |
| CN115953724B (en) | User data analysis and management method, device, equipment and storage medium | |
| CN117097425B (en) | Emergency scheme conflict detection method and system based on emergency broadcast system | |
| CN111861126B (en) | Assessment method, device and equipment for teaching plan | |
| CN111597357B (en) | Evaluation system and method for foundation learning |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190719 |
|
| WD01 | Invention patent application deemed withdrawn after publication |