CN109256114A - Phonic warning method for power system monitor scheduling - Google Patents
Phonic warning method for power system monitor scheduling Download PDFInfo
- Publication number
- CN109256114A CN109256114A CN201811003840.8A CN201811003840A CN109256114A CN 109256114 A CN109256114 A CN 109256114A CN 201811003840 A CN201811003840 A CN 201811003840A CN 109256114 A CN109256114 A CN 109256114A
- Authority
- CN
- China
- Prior art keywords
- alarm
- text
- facility information
- voice
- warning method
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L13/00—Speech synthesis; Text to speech systems
- G10L13/08—Text analysis or generation of parameters for speech synthesis out of text, e.g. grapheme to phoneme translation, prosody generation or stress or intonation determination
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B21/00—Alarms responsive to a single specified undesired or abnormal condition and not otherwise provided for
- G08B21/18—Status alarms
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/26—Speech to text systems
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/03—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters
- G10L25/18—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the type of extracted parameters the extracted parameters being spectral information of each sub-band
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L25/00—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00
- G10L25/27—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique
- G10L25/30—Speech or voice analysis techniques not restricted to a single one of groups G10L15/00 - G10L21/00 characterised by the analysis technique using neural networks
Abstract
The present invention provides the phonic warning method dispatched for power system monitor, belong to scheduling field, including facility information is obtained based on monitoring business interface;The facility information received is identified according to data type difference;Alarm voice is generated to the facility information for the accident that is identified as, and the alarm voice of generation is sent to corresponding treatment people.By introducing virtual customer service, realizes the real combination of technology and business, dispatcher is allowed to revert in prior work, play bigger value, alleviate growing basic staff demand pressure, while improving scheduling prison screen operating efficiency, modular working process.Outgoing call tell by telephone dispatcher or scene operation operating personnel.Dispatcher and on-site personnel are according to the warning content received, confirmation feedback is carried out, and executes relevant operation, for some special circumstances, related personnel informs simultaneously with other multimedia forms such as short message, picture in intelligent Jian Ping robot, realizes experience effect more preferably human-computer interaction.
Description
Technical field
The invention belongs to scheduling fields, in particular to the phonic warning method for power system monitor scheduling.
Background technique
With the fast development of domestic city process, the construction of urban power distribution network also carries out in high gear.In turn
Cause power distribution network radiating surface more and more wider, structure becomes increasingly complex, and asset size exponentially increases, to dispatcher and live work
Cheng personnel bring huge pressure.At certain periods (such as during summer peak meeting), dispatch all kinds of alarms on long-range monitoring large-size screen monitors,
The signals such as out-of-limit, exception, displacement are emerged in large numbers in " explosive ", need to expend a large amount of manpower carry out warning information reading analysis,
Assign and follow up and dispose, efficiency is relatively low, and is really worth without playing dispatcher.
In recent years, it as artificial intelligence technology continues to develop, all begins trying to replace people with AI technology in many fields
Work handles certain law, repeatability, business and is worth not high work, on the one hand can alleviate personnel's input cost, another party
People can be freed from the relatively low work of value again and play bigger value by face.Currently, dispatching of power netwoks prison screen work is main
Or manpower is relied on, it is on duty by dispatcher 24 hours, to handle all kinds of alarm signals of discovery.
Summary of the invention
In order to solve shortcoming and defect existing in the prior art, the present invention provides the voices dispatched for power system monitor
Alarm method can alleviate growing basic staff demand pressure, while improve scheduling prison screen operating efficiency, modular working
Process.
It is described the present invention provides the phonic warning method dispatched for power system monitor in order to reach above-mentioned technical purpose
Phonic warning method includes:
Obtained based on monitoring business interface includes alarm signal data, equipment account data, equipment deficiency data, Ren Yuanlian
It is the facility information including mode;
The facility information received is identified according to data type difference;
Alarm voice is generated to the facility information for the accident that is identified as, and the alarm voice of generation is sent to corresponding processing
Personnel.
Optionally, the described pair of facility information received is identified according to data type difference, comprising:
Facility information including protection act, emergency stop valve trip, ground fault is labeled as trouble-signal;
The facility information stored in OMS defect library will be belonged to labeled as abnormal signal;
The main transformer switch of D5000 equipment library storage and the facility information of line switching displacement will be belonged to labeled as abnormal letter
Number;
To belong to can not have the facility information mark that voltage and System Reactive Power are adjusted when AVC system runs abnormal
It is denoted as out-of-limit signal.
Optionally, the described pair of facility information for being identified as accident generates alarm voice, comprising:
Alarm text is determined according to the facility information labeled as accident;
By neural network speech synthesis technique, alarm voice corresponding with alarm text is obtained.
Optionally, the basis determines alarm text labeled as the facility information of accident, comprising:
Content planning is carried out to warning content, the warning content and structure to be generated are determined and are constructed;
Based on the details for explicitly defining planning text in Microplanner, carry out that word and optimization polymerization is selected to generate expression formula;
Mainly the structure and language that generate text are realized in Surface Realize, that is, will be after Microplanner
Text description map to the alarm text being made of text, punctuation mark and structure annotating information.
Optionally, described by neural network speech synthesis technique, obtain alarm voice corresponding with alarm text, packet
It includes:
It determines alarm switching strategy of the text to voice, neural network is trained based on fixed switching strategy,
Obtain source-target switching network of each syllable;
Alarm text is converted based on the switching network reached, obtains alarm voice corresponding with alarm text.
Optionally, the determining alarm switching strategy of the text to voice, comprising:
The parameter LSF for choosing the expression frequency spectrum of current mainstream carries out Speech processing, obtains spectrum distribution situation;
It the use of syllable is that unit carries out mapping network Construction of A Model according to existing corpus scale;
The processing that each parameter that will enter into neural network is normalized;
When deep neural network is into after crossing non-linear conversion, selection is turned with the immediate 30 groups of data of target component
It changes, when the parameter of conversion, which is not met, to sort by rank, just directly replaces the parameter with the nearest vector of distance in target corpus.
Optionally, described that alarm text is converted based on the switching network reached, it obtains corresponding with alarm text
Alert voice, comprising:
Corresponding deep neural network is picked out for converting by the markup information of parameter to be converted, while picking out synthesis
The frequency spectrum of corpus and original language material, for unified normalized;
Frequency spectrum to be converted after normalization is just used as the input parameter of corresponding conversion network, exports by deep neural network
Frequency spectrum after being converted;
Judge whether conversion spectrum has the characteristic arranged by rank, substitutes this with the frequency spectrum of original language material if not having
Frame finally obtains stable conversion spectrum, synthesizes to obtain alarm voice finally by filter.
Technical solution provided by the invention has the benefit that
By introducing virtual customer service, realizes the real combination of technology and business, dispatcher is allowed to revert to prior work
On work, bigger value is played, growing basic staff demand pressure is alleviated, while improving scheduling prison screen operating efficiency,
Modular working process.Outgoing call tell by telephone dispatcher or scene operation operating personnel.Dispatcher and on-site personnel are according to receipts
The warning content arrived carries out confirmation feedback, and executes relevant operation, and for some special circumstances, intelligent Jian Ping robot is simultaneously
Related personnel is informed with other multimedia forms such as short message, picture, realizes experience effect more preferably human-computer interaction.
Detailed description of the invention
It, below will be to attached drawing needed in embodiment description in order to illustrate more clearly of technical solution of the present invention
It is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, general for this field
For logical technical staff, without creative efforts, it is also possible to obtain other drawings based on these drawings.
Fig. 1 is the flow diagram provided by the present invention for the phonic warning method of power system monitor scheduling.
Specific embodiment
To keep structure and advantage of the invention clearer, structure of the invention is made further below in conjunction with attached drawing
Description.
Embodiment one
The present invention provides the phonic warning methods dispatched for power system monitor, as shown in Figure 1, the phonic warning method
Include:
11, being obtained based on monitoring business interface includes alarm signal data, equipment account data, equipment deficiency data, people
Facility information including member's contact method;
12, the facility information received is identified according to data type difference;
13, alarm voice is generated to the facility information of the accident that is identified as, and the alarm voice of generation is sent to corresponding
Treatment people.
In an implementation, the phonic warning method that the present embodiment proposes, can be realized distribution scheduling signal processing, analyzes and examine
It is disconnected, and then the alarm field of core can be exported.Specifically by being trained to business scene corpus, building is supervised towards scheduling
Shield the spatial term model of scene, the alarm field based on structuring can automatically generate the dialogue of natural language form
Text.
During realizing speech production, it is based on deep neural network speech synthesis technique, realizes alarm content of text
Speech synthesis is completed to converse with other operating personnels.It will can should remotely be supervised by using text generation and speech synthesis technique
It surveys and analyzes result and report in the form of phone, text, picture etc. are multimedia to relevant staff.
In addition the phonic warning method that the present embodiment proposes, is with virtual customer service or intelligence machine in actual use
The form of people embodies.
The intelligent robot system that the present embodiment is related to is related to four layers
(1) business original system layer: as the data source of AI engine, relevant business datum is mainly provided and such as alerts letter
Number information, equipment account information, equipment deficiency information etc. judge and are analyzed for business;
(2) artificial intelligence engine layer: as the nucleus module of intelligent Jian Ping robot, in addition to AI base engines, while
On artificial intelligence technology, all kinds of AI application engines of the training building towards distribution scheduling field under the guidance of business expert;
(3) all kinds of AI engine services AI engine calling service layer: are provided for the application function on upper layer;
(4) application function layer: the virtual monitor robot for having business diagnosis and making and receiving calls function is provided.
By introducing the virtual customer service in forward position, the real combination of technology and business is realized;Allow scheduling for social perspective
Personnel revert in prior work, play bigger value, alleviate growing basic staff demand pressure, mention simultaneously
Scheduling prison screen operating efficiency, modular working process;Substantially 90% or more warning information artificial treatment work can be taken over, greatly
The big labor intensity for mitigating artificial prison screen work.And by the telephone system of deployment, and the personal information obtained, outer calling telephone
Inform dispatcher or scene operation operating personnel.Dispatcher and on-site personnel are confirmed according to the warning content received
Feedback, and relevant operation is executed, for some special circumstances, intelligent Jian Ping robot is simultaneously with other more matchmakers such as short message, picture
Body form informs related personnel, realizes experience effect more preferably human-computer interaction.
Specifically, what is referred in step 12 is identified the facility information received according to data type difference, comprising:
121, the facility information including protection act, emergency stop valve trip, ground fault is labeled as trouble-signal;
122, the facility information stored in OMS defect library will be belonged to labeled as abnormal signal;
123, the main transformer switch of D5000 equipment library storage and the facility information of line switching displacement will be belonged to labeled as abnormal
Signal;
124, it will belong to exist when AVC system runs abnormal and the equipment that voltage and System Reactive Power are adjusted believed
Breath is labeled as out-of-limit signal.
In an implementation, business datum is obtained by the data-interface that operation system is developed, according to the alarm data obtained
Different alarm types is identified with known business rule, current processing logic is as follows:
Trouble-signal includes mainly the signals such as protection act, emergency stop valve trip, ground fault, directly equipment should be notified to run
Unit personnel inspection, and inform as value regulation person;
Abnormal signal, compares OMS defect library, and signal belongs to defect processing skipping automatically on way, is not belonging to handle on way
Notice equipment run unit personnel inspection, and inform when value regulation person;
Signal is conjugated, D5000 equipment library is compareed, is not belonging to (capacitor, reactor of main transformer switch and line switching displacement
Switching and main shift are adjusted) it skips automatically, the notice equipment run unit personnel inspection belonged to, and inform as value regulation person;
Out-of-limit signal, mainly for two class of voltage out-of-limit and reactive power constraints, under normal circumstances, AVC system can to its into
Row automatic adjustment, but when the operation of AVC system is abnormal, exist and accurate and effective adjusting not can be carried out to voltage and System Reactive Power
The case where.In intelligence prison screen, intelligent robot should track out-of-limit signal, continue more limit and be greater than the set value (delay
Processing) after, notice regulation person carries out manual adjustment.
Optionally, the facility information to the accident that is identified as being related in step 13 generates alarm voice, comprising:
131, alarm text is determined according to the facility information labeled as accident;
132, by neural network speech synthesis technique, alarm voice corresponding with alarm text is obtained.
Wherein step 131 includes:
1311, content planning is carried out to warning content, the warning content and structure to be generated is determined and is constructed;
1312, it based on the details for explicitly defining planning text in Microplanner, carries out that word and optimization polymerization is selected to generate expression
Formula;
1313, mainly the structure and language that generate text are realized in Surface Realize, that is, will be through microcosmic
Text description after planning maps to the alarm text being made of text, punctuation mark and structure annotating information.
In an implementation, accident, abnormal, out-of-limit and four class alarm signals of displacement are mainly supervised by intelligent Jian Ping robot
Control.Based on the signal data and account data that get and defect information etc., it then follows the semanteme and rule of scheduling prison frequency business,
Corresponding alarm speech text is automatically generated according to different alarm types.With pipeline model classical in NLG, have preferable
Robustness, independence and reusability.It is specific as follows to alert speech text generation step:
1) warning content and structure to be generated are determined and are constructed by content planning.
2) details that planning text is explicitly defined in Microplanner such as the account information of alarm equipment, location information, lacks
Information etc. is fallen into, needs further to select word and optimization polymerization, then submit generation expression formula.Present apparatus institute's research contents generates expression formula
(template) is exemplified below:
XX O&M class, XX points when XX, XX, which becomes XX signal, to be occurred, and the scene that please arrive is checked;
XX power supply station, XX points when XX, XX line protection act, switch trip, successful reclosing (or failure), XX points when XX, perhaps
It can accident line walking;
As value regulation person, XX, which becomes XX signal, to be occurred, and XX O&M class has been notified to check;
As value regulation person, XX line protection act, switch trip, successful reclosing (or failure) notified XX O&M class to check,
And permit XX accident line walking;
As value regulation person, XX time variant voltage is out-of-limit, please adjusts in time.
…
It is microcosmic by selecting word, polymerization and generation expression formula being submitted to complete in this step with reference to pipeline model
Content planning.
A) it selects word: selecting word to need to consider context environmental, I-goal and practical factor etc. in the application, in the present apparatus
The word of selection need to meet scheduling prison screen service scene, such as device name, alarm type, and it is daily that syntactic structure will meet dispatcher
Dialogue habit;
B) it polymerize: eliminates the redundancy between sentence, increase the readability of text, the syntactic structure of warning content has one
As have the call format of standard, i.e. using directly carrying out group to excessively a short sentence using conjunction by the way of simply connecting
It closes.
C) it submits and generates expression formula: in realization after the step of two, face, submitting generating expression formula and mainly allow warning content
Expression have more language modifying, such as reduce duplicate reference, increase can sentence entirety readability.
3) mainly the structure and language that generate text are realized in Surface Realize, that is, will be through Microplanner
Text description afterwards maps to the surface layer text being made of text, punctuation mark and structure annotating information.
Content of text finally to be alerted is formed by these three above big steps.
Optionally, step 132 includes:
1321, it determines alarm switching strategy of the text to voice, neural network is carried out based on fixed switching strategy
Training, obtains source-target switching network of each syllable;
1322, alarm text is converted based on the switching network reached, obtains alarm language corresponding with alarm text
Sound.
What step 1311 was mentioned to determines alarm switching strategy of the text to voice, comprising:
The parameter LSF for choosing the expression frequency spectrum of current mainstream carries out Speech processing, obtains spectrum distribution situation;
It the use of syllable is that unit carries out mapping network Construction of A Model according to existing corpus scale;
The processing that each parameter that will enter into neural network is normalized;
When deep neural network is into after crossing non-linear conversion, selection is turned with the immediate 30 groups of data of target component
It changes, when the parameter of conversion, which is not met, to sort by rank, just directly replaces the parameter with the nearest vector of distance in target corpus.
In an implementation, by the markup information of parameter to be converted, identical syllable is chosen from synthesis voice and original language material
The frequency spectrum parameter of parallel corpora, is uniformly normalized after time unifying, obtained normalized parameter respectively as
The input and output parameter of deep neural network is learnt.Obtain source-target switching network of each syllable.
Corresponding step 1312, comprising:
Corresponding deep neural network is picked out for converting by the markup information of parameter to be converted, while picking out synthesis
The frequency spectrum of corpus and original language material, for unified normalized;
Frequency spectrum to be converted after normalization is just used as the input parameter of corresponding conversion network, exports by deep neural network
Frequency spectrum after being converted;
Judge whether conversion spectrum has the characteristic arranged by rank, substitutes this with the frequency spectrum of original language material if not having
Frame finally obtains stable conversion spectrum, synthesizes to obtain alarm voice finally by filter.
In an implementation, corresponding deep neural network is picked out for converting, simultaneously by the markup information of parameter to be converted
The frequency spectrum for picking out synthesis corpus and original language material, for unified normalized.Frequency spectrum to be converted after normalization is just made
For the input parameter of corresponding conversion network, the frequency spectrum after being converted is exported by deep neural network.Judging conversion spectrum is
The no characteristic having by rank arrangement, substitutes this frame with the frequency spectrum of original language material if not having, finally obtains stable conversion
Frequency spectrum is synthesized finally by filter.In this way, can also complete training under conditions of limited training parameter, convert and close
At the step of achieve the effect that improve sound quality.
The present invention provides the phonic warning methods dispatched for power system monitor, including are set based on the acquisition of monitoring business interface
Standby information;The facility information received is identified according to data type difference;The facility information for the accident that is identified as is generated
Voice is alerted, and the alarm voice of generation is sent to corresponding treatment people.By introducing virtual customer service, technology and industry are realized
The real combination of business, allows dispatcher to revert in prior work, plays bigger value, alleviates growing base
Personnel demand pressure, while improving scheduling prison screen operating efficiency, modular working process.Outgoing call tell by telephone dispatcher or scene fortune
Row operating personnel.Dispatcher and on-site personnel carry out confirmation feedback according to the warning content received, and execute related behaviour
Make, for some special circumstances, relevant people is informed simultaneously with other multimedia forms such as short message, picture by intelligent Jian Ping robot
Member realizes experience effect more preferably human-computer interaction.
Each serial number in above-described embodiment is for illustration only, the assembling for not representing each component or the elder generation in use process
Sequence afterwards.
The above description is only an embodiment of the present invention, is not intended to limit the invention, all in the spirit and principles in the present invention
Within, any modification, equivalent replacement, improvement and so on should all be included in the protection scope of the present invention.
Claims (7)
1. the phonic warning method for power system monitor scheduling, which is characterized in that the phonic warning method includes:
Obtained based on monitoring business interface includes alarm signal data, equipment account data, equipment deficiency data, personnel's correspondent party
Facility information including formula;
The facility information received is identified according to data type difference;
Alarm voice is generated to the facility information for the accident that is identified as, and the alarm voice of generation is sent to corresponding processing people
Member.
2. the phonic warning method according to claim 1 for power system monitor scheduling, which is characterized in that described pair of reception
To facility information be identified according to data type difference, comprising:
Facility information including protection act, emergency stop valve trip, ground fault is labeled as trouble-signal;
The facility information stored in OMS defect library will be belonged to labeled as abnormal signal;
The main transformer switch of D5000 equipment library storage and the facility information of line switching displacement will be belonged to labeled as abnormal signal;
To belong to when AVC system runs abnormal, which can not have the facility information that voltage and System Reactive Power are adjusted, is labeled as
Out-of-limit signal.
3. the phonic warning method according to claim 1 for power system monitor scheduling, which is characterized in that described pair of mark
Alarm voice is generated for the facility information of accident, comprising:
Alarm text is determined according to the facility information labeled as accident;
By neural network speech synthesis technique, alarm voice corresponding with alarm text is obtained.
4. the phonic warning method according to claim 3 for power system monitor scheduling, which is characterized in that described according to mark
The facility information for being denoted as accident determines alarm text, comprising:
Content planning is carried out to warning content, the warning content and structure to be generated are determined and are constructed;
Based on the details for explicitly defining planning text in Microplanner, carry out that word and optimization polymerization is selected to generate expression formula;
Mainly the structure and language that generate text are realized in Surface Realize, that is, by the text after Microplanner
This description maps to the alarm text being made of text, punctuation mark and structure annotating information.
5. the phonic warning method according to claim 3 for power system monitor scheduling, which is characterized in that described by mind
Through voice-over-net synthetic technology, alarm voice corresponding with alarm text is obtained, comprising:
It determines alarm switching strategy of the text to voice, neural network is trained based on fixed switching strategy, is obtained
The source of each syllable-target switching network;
Alarm text is converted based on the switching network reached, obtains alarm voice corresponding with alarm text.
6. the phonic warning method according to claim 5 for power system monitor scheduling, which is characterized in that the determining announcement
Alert switching strategy of the text to voice, comprising:
The parameter LSF for choosing the expression frequency spectrum of current mainstream carries out Speech processing, obtains spectrum distribution situation;
It the use of syllable is that unit carries out mapping network Construction of A Model according to existing corpus scale;
The processing that each parameter that will enter into neural network is normalized;
When deep neural network is into after crossing non-linear conversion, selection is converted with the immediate 30 groups of data of target component, when
The parameter of conversion is not met when sorting by rank, just directly replaces the parameter with the nearest vector of distance in target corpus.
7. the phonic warning method according to claim 6 for power system monitor scheduling, which is characterized in that described to be based on reaching
To switching network alarm text is converted, obtain and alert the corresponding alarm voice of text, comprising:
Corresponding deep neural network is picked out for converting by the markup information of parameter to be converted, while picking out synthesis corpus
With the frequency spectrum of original language material, for unified normalized;
Frequency spectrum to be converted after normalization is just used as the input parameter of corresponding conversion network, exports to obtain by deep neural network
Frequency spectrum after conversion;
Judge whether conversion spectrum has the characteristic arranged by rank, substitute this frame with the frequency spectrum of original language material if not having,
Stable conversion spectrum is finally obtained, synthesizes to obtain alarm voice finally by filter.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811003840.8A CN109256114B (en) | 2018-08-30 | 2018-08-30 | Voice alarm method for power grid monitoring and dispatching |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811003840.8A CN109256114B (en) | 2018-08-30 | 2018-08-30 | Voice alarm method for power grid monitoring and dispatching |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109256114A true CN109256114A (en) | 2019-01-22 |
CN109256114B CN109256114B (en) | 2022-12-13 |
Family
ID=65049815
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811003840.8A Active CN109256114B (en) | 2018-08-30 | 2018-08-30 | Voice alarm method for power grid monitoring and dispatching |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109256114B (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111128144A (en) * | 2019-10-16 | 2020-05-08 | 国网浙江省电力有限公司金华供电公司 | Voice power grid dispatching system and method |
CN112468668A (en) * | 2020-11-24 | 2021-03-09 | 国网河南省电力公司南阳供电公司 | Telephone alarming method and system of power dispatching monitoring system |
CN113205799A (en) * | 2021-03-24 | 2021-08-03 | 合肥佳讯科技有限公司 | Alarm processing method based on voice recognition |
CN113447744A (en) * | 2021-06-28 | 2021-09-28 | 国网福建省电力有限公司福州供电公司 | Equipment abnormity alarm event synthesis method of power monitoring system based on signal analysis |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102750795A (en) * | 2012-06-21 | 2012-10-24 | 江苏省电力公司苏州供电公司 | Acousto-optic alarm device |
CN104811327A (en) * | 2014-01-26 | 2015-07-29 | 中国移动通信集团江西有限公司 | Monitoring warning voice automatic notification method and device |
WO2016029570A1 (en) * | 2014-08-28 | 2016-03-03 | 北京科东电力控制系统有限责任公司 | Intelligent alert analysis method for power grid scheduling |
-
2018
- 2018-08-30 CN CN201811003840.8A patent/CN109256114B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102750795A (en) * | 2012-06-21 | 2012-10-24 | 江苏省电力公司苏州供电公司 | Acousto-optic alarm device |
CN104811327A (en) * | 2014-01-26 | 2015-07-29 | 中国移动通信集团江西有限公司 | Monitoring warning voice automatic notification method and device |
WO2016029570A1 (en) * | 2014-08-28 | 2016-03-03 | 北京科东电力控制系统有限责任公司 | Intelligent alert analysis method for power grid scheduling |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111128144A (en) * | 2019-10-16 | 2020-05-08 | 国网浙江省电力有限公司金华供电公司 | Voice power grid dispatching system and method |
CN112468668A (en) * | 2020-11-24 | 2021-03-09 | 国网河南省电力公司南阳供电公司 | Telephone alarming method and system of power dispatching monitoring system |
CN113205799A (en) * | 2021-03-24 | 2021-08-03 | 合肥佳讯科技有限公司 | Alarm processing method based on voice recognition |
CN113447744A (en) * | 2021-06-28 | 2021-09-28 | 国网福建省电力有限公司福州供电公司 | Equipment abnormity alarm event synthesis method of power monitoring system based on signal analysis |
Also Published As
Publication number | Publication date |
---|---|
CN109256114B (en) | 2022-12-13 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109256114A (en) | Phonic warning method for power system monitor scheduling | |
CN207319184U (en) | A kind of field device maintenance maintenance and current check system | |
CN111325323A (en) | Power transmission and transformation scene description automatic generation method fusing global information and local information | |
CN110188331A (en) | Model training method, conversational system evaluation method, device, equipment and storage medium | |
CN111755001A (en) | Artificial intelligence-based power grid rapid dispatching and commanding system and method | |
CN108629432A (en) | Troubleshooting planing method, system and device | |
CN109345072A (en) | Dispatching method for distribution branch line scheduling scenario | |
CN112599124A (en) | Voice scheduling method and system for power grid scheduling | |
CN209374053U (en) | Dispatching intelligent monitor and check system | |
CN110046232A (en) | Natural expression processing method, response method, equipment and the system of natural intelligence | |
CN110059166A (en) | Natural expression processing method, response method, equipment and the system of natural intelligence | |
CN111384778B (en) | Intelligent operation and maintenance system for power distribution network equipment | |
CN109284386A (en) | Customized intension recognizing method and device | |
CN111179928A (en) | Intelligent control method for power transformation and distribution station based on voice interaction | |
CN110059168A (en) | The method that man-machine interactive system based on natural intelligence is trained | |
CN110096516A (en) | The dialogue generation method and system of customized database interaction | |
CN116468287A (en) | Intelligent park control system based on digital twinning | |
CN106295992A (en) | Wind power generating set condition monitoring based on hologram and O&M instruct system | |
CN116542241A (en) | Matching method of emergency plan and emergency medical rescue cooperative command platform system | |
CN110059167A (en) | Natural expression processing method, response method, equipment and the system of natural intelligence | |
CN110008317A (en) | Natural expression processing method, response method, equipment and the system of natural intelligence | |
CN116804691B (en) | Fault monitoring method for dispatching automation equipment of power system | |
CN116563499A (en) | Intelligent interaction system of transformer substation based on meta-universe technology | |
CN113946464B (en) | Alarm noise reduction method combining model and experience pre-training and parallel deduction | |
Xiang et al. | Design of Intelligent Dispatching System Based on Human Voice Adaptive Speech Recognition |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |