CN114936801A - Distribution network dispatching operation management method based on big data - Google Patents
Distribution network dispatching operation management method based on big data Download PDFInfo
- Publication number
- CN114936801A CN114936801A CN202210689002.0A CN202210689002A CN114936801A CN 114936801 A CN114936801 A CN 114936801A CN 202210689002 A CN202210689002 A CN 202210689002A CN 114936801 A CN114936801 A CN 114936801A
- Authority
- CN
- China
- Prior art keywords
- distribution network
- data
- service
- implementation
- abnormal
- 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.)
- Withdrawn
Links
- 238000007726 management method Methods 0.000 title claims abstract description 35
- 230000002159 abnormal effect Effects 0.000 claims abstract description 52
- 230000000694 effects Effects 0.000 claims abstract description 17
- 238000012795 verification Methods 0.000 claims abstract description 17
- 238000012549 training Methods 0.000 claims description 53
- 238000004458 analytical method Methods 0.000 claims description 40
- 230000008447 perception Effects 0.000 claims description 33
- 230000009466 transformation Effects 0.000 claims description 14
- 238000011156 evaluation Methods 0.000 claims description 12
- 238000012360 testing method Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000003062 neural network model Methods 0.000 claims description 10
- 238000012216 screening Methods 0.000 claims description 9
- 230000005856 abnormality Effects 0.000 claims description 6
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 10
- 238000012545 processing Methods 0.000 abstract description 10
- 238000000034 method Methods 0.000 abstract description 7
- 238000002372 labelling Methods 0.000 abstract description 5
- 238000012544 monitoring process Methods 0.000 description 4
- 230000010354 integration Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06312—Adjustment or analysis of established resource schedule, e.g. resource or task levelling, or dynamic rescheduling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/36—Creation of semantic tools, e.g. ontology or thesauri
- G06F16/367—Ontology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
- G06Q10/06313—Resource planning in a project environment
-
- 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Entrepreneurship & Innovation (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- General Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- General Business, Economics & Management (AREA)
- Mathematical Physics (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Molecular Biology (AREA)
- Tourism & Hospitality (AREA)
- Biophysics (AREA)
- Marketing (AREA)
- Biomedical Technology (AREA)
- Artificial Intelligence (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Development Economics (AREA)
- Biodiversity & Conservation Biology (AREA)
- Animal Behavior & Ethology (AREA)
- Databases & Information Systems (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The invention discloses a distribution network dispatching operation management method based on big data, which belongs to the technical field of distribution network dispatching operation management, and is characterized in that the data of each service during the operation of a distribution network are standardized and normalized and matched with a check list pre-stored in a database to judge whether the plan implementation of the corresponding service is normal or not, then the judgment results of all the services are integrated in a simultaneous manner, and the implementation degree is obtained through calculation to evaluate the plan implementation condition of the whole service, so that the abnormity can be found in time and the processing can be carried out in a targeted manner, and the overall effect of the distribution network service plan implementation can be effectively improved; the method and the device are used for solving the technical problem that the overall effect of distribution network scheduling operation management is poor due to the fact that the overall situation of each service plan implementation of a distribution network cannot be monitored and evaluated and abnormal data of different types of services are subjected to labeling processing and verification in the existing scheme.
Description
Technical Field
The invention relates to the technical field of distribution network dispatching operation management, in particular to a distribution network dispatching operation management method based on big data.
Background
Because the distribution network has the characteristics of numerous equipment, rapid grid structure change, wide involvement range and the like, certain difficulty still exists in the implementation of the current power grid dispatching management, most power departments usually only pay attention to the dispatching management of a main network, and ignore the dispatching management of the distribution network, so that the overall implementation of the current power distribution network dispatching management can be limited, the standardized management level of the power distribution network dispatching is improved, better technical support is provided for the power distribution operation, and the problem to be solved urgently in the current power industry is solved.
When the existing distribution network scheduling operation management scheme is implemented, the implementation conditions of different service plans cannot be automatically processed, and the results of all service processing are simultaneously integrated to integrally evaluate, so that the plan implementation of each service is mutually isolated; moreover, labeling processing and verification are not performed on the monitored abnormal data, and overall monitoring and studying and judging are not performed on the perception of the distribution network, so that the overall effect of distribution network scheduling operation management is poor.
Disclosure of Invention
The invention aims to provide a distribution network scheduling operation management method based on big data, which solves the following technical problems: how to solve the technical problem that the overall effect of distribution network scheduling operation management is poor because the overall situation of each service plan implementation of a distribution network cannot be monitored and evaluated and abnormal data of different types of services are subjected to labeling processing and verification in the conventional scheme.
The distribution network scheduling operation management method based on big data comprises the following steps:
acquiring data of each service when a distribution network operates and combining the data to obtain a service operation set; the service operation set comprises an OMS power failure plan, transaction management, a scheduling log, distribution network protection information, distribution network graph model topology and equipment state, fault study and judgment data, acquisition stop and power supply, distribution transformer load data and self-terminal data;
checking the operation service set to evaluate whether the plan implementation of each service is abnormal or not to obtain a service analysis set, and evaluating the implementation state of the whole service according to the service analysis set to obtain an implementation evaluation result;
training the acquired running state data through a pre-constructed distribution network running abnormity analysis model, acquiring state labels corresponding to running states in the running state data, analyzing and screening the state labels to obtain normal labels and abnormal labels, and constructing a distribution network running abnormity knowledge map through the abnormal labels;
calculating data of each service early warning during the operation of the distribution network to obtain a corresponding sensing result, wherein the sensing result comprises a distribution transformer sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate;
and analyzing and checking the calculated sensing result in sequence to obtain a checking result and carrying out alarm prompt.
Further, checking the running service set to evaluate whether the plan implementation of each service is abnormal or not includes:
sequentially numbering the service items in the service operation set and marking the service items as i, wherein the i is a positive integer;
sequentially extracting each text and number in each service item, and arranging and combining the texts and the numbers to obtain corresponding text data and digital data;
sequentially setting text names in the text data as identification marks, and respectively setting text contents corresponding to the text names and numbers in the digital data as a first verification mark and a second verification mark;
sequentially matching the identification marks with a check list pre-stored in a database according to the serial numbers of the service items, and judging whether a first check mark and a second check mark associated with the identification marks are matched with sample contents and sample numbers in the check list or not; wherein the check list comprises a plurality of identification marks and associated sample content and sample numbers;
when the service items are matched with each other, generating a first matching signal and setting the service item corresponding to the identification mark as a normal item;
when the identification marks are not matched with the service items, generating a second matching signal and setting the service items corresponding to the identification marks as target items;
the first and second matching signals, the normal item and the target item form a business analysis set.
Further, the evaluation of the implementation state of the whole service according to the service analysis set includes:
acquiring the total number of normal items and target items in a service analysis set, and respectively taking values and marking the values as ZZ and MZ;
matching the identification mark corresponding to the target item with a pre-constructed item name table to obtain a corresponding item weight and marking the item weight as XQ;
combining the normal items and the total number of the target items of the value marks and the item weights corresponding to the target items, and calculating and obtaining the implementation degree SS of the whole service through a formula; the formula is:
in the formula, a1 and a2 are different proportionality coefficients, and the value ranges are (0, 7);
matching the implementation degree with a preset implementation threshold value;
if the implementation degree is smaller than the implementation threshold value, judging that the implementation state of the whole service is excellent and generating a first implementation signal;
if the implementation degree is not less than the implementation threshold and not greater than m% of the implementation threshold, and m is a real number greater than one hundred, determining that the implementation state of the whole service is slightly abnormal and generating a second implementation signal;
if the implementation degree is larger than m% of the implementation threshold, judging that the implementation state of the whole service is moderate and abnormal, and generating a third implementation signal;
the first implementation signal, the second implementation signal and the third implementation signal constitute implementation evaluation results.
Further, the step of constructing the distribution network operation abnormity analysis model comprises the following steps:
acquiring a data set; the data set comprises a plurality of historical line state data, switch state data, distribution and transformation frequent fault data, corresponding power failure household data, power failure duration data and complaint data;
dividing the data set according to a preset division ratio to obtain a training set and a test set;
counting the total number N of the training sets, setting N of the training sets as a verification set, and taking the rest N-N as the training sets to be trained; n and N are positive integers, and N is more than N;
training the training set for N-N times through a neural network algorithm to obtain N-N different neural network models, wherein the neural network algorithm comprises an error reverse feedback neural network algorithm, an RBF neural network algorithm and a deep convolution neural network algorithm;
evaluating the effects of the N-N different neural network models by using the verification set, and screening out the hyper-parameters corresponding to the neural network model with the best effect;
the super-parameters with the best effect are used for retraining the model for the N training sets to obtain a distribution network operation abnormity analysis model;
and testing and evaluating the distribution network operation abnormity analysis model by using the test set.
Further, training the acquired running state data through a pre-constructed distribution network running abnormity analysis model, comprising:
acquiring a line state, a switch state and a distribution transformation frequent state in the running state data, training through a distribution network running abnormity analysis model, and acquiring corresponding training power failure number of users, training power failure duration and training complaint number;
and respectively matching the training power failure number of the users, the training power failure time length and the training complaint number with a preset power failure number table, a preset power failure time length table and a preset complaint table, and acquiring corresponding power failure number labels, power failure time length labels and complaint number labels.
Further, each status label is analyzed and screened, including:
matching the power failure household number label, the power failure duration label and the complaint number label with all label positions in a power failure household number table, a power failure duration table and a complaint table respectively;
if the matched power outage household number label, power outage duration label and complaint number label are positioned at the front p bits of all sequencing labels in the corresponding power outage household number table, power outage duration table and complaint table, and p is a positive integer, judging that the corresponding label is an abnormal label; otherwise, judging the label as a normal label;
and constructing a distribution network operation abnormal knowledge graph by using the abnormal labels and the corresponding states in the operation state data.
Further, calculating data of each service early warning during operation of the distribution network to obtain a corresponding perception result, including:
acquiring a distribution transformation total number, a switch total number, a transaction total number and a plan execution total number when the distribution network runs, and marking the distribution transformation total number, the switch total number, the transaction total number and the plan execution total number as YZi, wherein i is 1, 2, 3 and 4;
acquiring distribution transformer early warning number, switch early warning number, transaction early warning number and plan execution early warning number which occur when the distribution network operates, and marking the distribution transformer early warning number, the switch early warning number, the transaction early warning number and the plan execution early warning number as YYi;
carrying out simultaneous calculation on all marked data and obtaining a perception result GLi through formula calculation; the formula is:
GLi＝YYi/YZi
the sensing result comprises a distribution transformation sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate.
Further, analyzing and checking the calculated sensing result to obtain a checking result and performing alarm prompting, including:
acquiring a distribution variation constant, a switch differential constant, a differential motion constant and a plan execution differential constant of an approved distribution network during operation, and respectively combining the distribution variation constant, the switch differential constant, the differential motion constant and the plan execution differential constant with corresponding distribution total number, switch total number, differential motion total number and plan execution total number to obtain an approval result, wherein the approval result comprises a distribution variation approval rate, a switch approval rate, a differential motion approval rate and a plan execution approval rate;
obtaining a difference value between each core rate and the corresponding perception rate, and matching the difference value with the corresponding difference threshold value;
if the difference value is smaller than the difference threshold value, judging that the corresponding perception rate is effective and marking the corresponding operation type as effective perception;
if the difference value is not smaller than the difference threshold value, judging that the corresponding perception rate is invalid and marking the corresponding operation type as invalid perception;
and the difference value, the effective perception and the ineffective perception form a checking result.
Compared with the prior scheme, the invention has the following beneficial effects:
the method comprises the steps of standardizing and standardizing data of each service during the operation of the distribution network, matching the data with a check list pre-stored in a database to judge whether plan implementation of the corresponding service is normal or not, then performing simultaneous integration on judgment results of all the services, and evaluating plan implementation conditions of the whole service by calculating implementation degrees, so that abnormality can be found in time and processing can be performed pertinently, and the whole effect of plan implementation of the distribution network service can be effectively improved;
through simultaneous integration and classification of a plurality of historical data, support can be provided for subsequent analysis and evaluation of each state data, and through labeling and classification of abnormal state data, a distribution network operation abnormal knowledge map can be constructed for the abnormal labels and corresponding states in the operation state data, so that the operation rules and equipment characteristics of distribution networks in different areas can be mastered conveniently;
by calculating, analyzing and evaluating the early warning data of each service, the early warning perception condition corresponding to each abnormal service can be obtained in time, so that the abnormality and the abnormal degree can be found in time.
Drawings
The invention is further described below with reference to the accompanying drawings.
Fig. 1 is a flow chart of a distribution network scheduling operation management method based on big data according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Referring to fig. 1, the present invention is a distribution network scheduling operation management method based on big data, and the specific steps include:
s1: acquiring data of each service when a distribution network operates and combining the data to obtain a service operation set; the service operation set comprises an OMS power failure plan, transaction management, a scheduling log, distribution network protection information, distribution network graph model topology and equipment state, fault study and judgment data, acquisition stop and power supply, distribution transformer load data and self-terminal data;
the data needed by the operation and management of the power distribution network has the characteristics of wide range and large amount, and operators need a large amount of power distribution network information for decision support when performing service analysis such as alarm information analysis, fault diagnosis, user service and the like; in addition, each department and each professional concern are different, and data analysis and mining are required to be carried out on the relevance and the data rule of various types of data of the distribution network, so that the monitoring data of the operation of the distribution network are perfected, and the dispatching operation management level of the distribution network is further improved.
S2: checking the operation service set to evaluate whether the plan implementation of each service is abnormal or not to obtain a service analysis set, and evaluating the implementation state of the whole service according to the service analysis set to obtain an implementation evaluation result;
the checking and evaluating whether the plan implementation of each service is abnormal or not for the operation service set comprises the following steps:
sequentially numbering the service items in the service operation set and marking the service items as i, wherein the i is a positive integer;
sequentially extracting each text and number in each service item, and arranging and combining the texts and the numbers to obtain corresponding text data and digital data;
sequentially setting text names in the text data as identification marks, and respectively setting text contents corresponding to the text names and numbers in the digital data as a first verification mark and a second verification mark;
according to the serial number of the service item, matching the identification mark with a check list pre-stored in a database in sequence, and judging whether a first check mark and a second check mark associated with the identification mark are matched with sample content and sample numbers in the check list or not; wherein, the checking list comprises a plurality of identification marks and associated sample contents and sample numbers;
when the service items are matched with each other, generating a first matching signal and setting the service item corresponding to the identification mark as a normal item;
when the identification marks are not matched with the service items, generating a second matching signal and setting the service items corresponding to the identification marks as target items;
a plurality of first matching signals, second matching signals, normal items and target items form a business analysis set;
wherein, whether the plan is normally executed includes but is not limited to whether the power failure range is expanded or not, whether the abnormal content is consistent with the abnormal content, whether the changing and abnormal operation are frequently carried out, whether the log record exists when the equipment is in failure stop or not, whether the position plate of the image model equipment meets the time requirement or not, whether the maintenance plate is still loaded data when the equipment is hung and stopped, and whether the failure research and judgment data is accurate or not;
evaluating the implementation state of the whole service according to the service analysis set, comprising:
acquiring the total number of normal items and target items in a service analysis set, and respectively taking values and marking the values as ZZ and MZ;
matching the identification mark corresponding to the target item with a pre-constructed item name table to obtain a corresponding item weight and marking the item weight as XQ;
the project name table comprises a plurality of different identification marks and corresponding project weights thereof, and the different identification marks are preset with one corresponding project weight; different types of services correspond to different importance, and differentiation and digital representation of the services are realized based on the project weight;
combining the normal items and the total number of the target items of the value marks and the item weights corresponding to the target items, and calculating and obtaining the implementation degree SS of the whole service through a formula; the formula is:
in the formula, a1 and a2 are different proportionality coefficients, the value ranges are (0,7), a1 can be 1.362, and a2 can be 3.483;
the implementation degree is a numerical value for integrally evaluating whether there is an abnormality in planned implementation of each service; in the process of calculating the implementation degree, the smaller the number of the target items, the smaller the item weight corresponding to the target items, the smaller the implementation degree obtained by calculation, and the more excellent the planned implementation state corresponding to the overall business.
Matching the implementation degree with a preset implementation threshold value;
if the implementation degree is smaller than the implementation threshold value, judging that the implementation state of the whole service is excellent and generating a first implementation signal;
if the implementation degree is not less than the implementation threshold and not greater than m% of the implementation threshold, m is a real number greater than one hundred, and can be 130, determining that the implementation state of the whole service is slightly abnormal, and generating a second implementation signal;
if the implementation degree is larger than m% of the implementation threshold, judging that the implementation state of the whole service is moderate and abnormal, and generating a third implementation signal;
the first implementation signal, the second implementation signal and the third implementation signal constitute implementation evaluation results.
In the embodiment of the invention, the data of each service during the operation of the distribution network is standardized and normalized, and is matched with the check list pre-stored in the database to judge whether the plan implementation of the corresponding service is normal or not, then the judgment results of all the services are integrated in a simultaneous manner, and the implementation degree is obtained through calculation to evaluate the plan implementation condition of the whole service, so that the abnormality can be found in time and the processing can be carried out in a targeted manner, and the overall effect of the plan implementation of the distribution network service can be effectively improved.
S3: training the acquired running state data through a pre-constructed distribution network running abnormity analysis model, acquiring state labels corresponding to all running states in the running state data, analyzing and screening all the state labels to obtain normal labels and abnormal labels, and constructing a distribution network running abnormity knowledge graph through the abnormal labels;
the method comprises the following steps of constructing a distribution network operation abnormity analysis model:
acquiring a data set; the data set comprises a plurality of historical line state data, switch state data, distribution transformation frequent fault data, corresponding power failure household number data, power failure duration data and complaint data; the unit of the power failure time is minutes; various text data in the historical line state data, switch state data and distribution and transformation frequent fault data can be subjected to digital processing, for example, value assignment is carried out according to the type corresponding to the text data; the same processing can be carried out on various text data acquired subsequently;
dividing the data set according to a preset division ratio to obtain a training set and a test set;
counting the total number N of the training sets, setting N of the training sets as a verification set, and taking the rest N-N as the training sets to be trained; n and N are positive integers, and N is more than N; the training set is used for training the model, the verification set is used for adjusting and selecting the model, and the test set is used for evaluating the final model;
training the training set for N-N times through a neural network algorithm to obtain N-N different neural network models, wherein the neural network algorithm comprises an error reverse feedback neural network algorithm, an RBF neural network algorithm and a deep convolution neural network algorithm;
evaluating the effects of N-N different neural network models by using the verification set, and screening out the hyper-parameters corresponding to the neural network model with the best effect; the effect here may preferably be that the error is minimal;
the super-parameters with the best effect are used for retraining the model for the N training sets to obtain a distribution network operation abnormity analysis model;
testing and evaluating the distribution network operation abnormity analysis model by using the test set;
in the embodiment of the invention, the data set division, the training set training, the evaluation of the effect of the neural network model and the test evaluation by using the test set are all the existing conventional technical means, and the specific steps are not repeated herein; in addition, the purpose of constructing a distribution network operation abnormity analysis model is to identify and classify distribution network operation state data to obtain corresponding labels so as to provide data support for the subsequent construction of a distribution network operation abnormity knowledge graph;
the method comprises the following steps of training acquired running state data through a pre-constructed distribution network running abnormity analysis model, wherein the training comprises the following steps:
acquiring a line state, a switch state and a distribution transformation frequent state in the running state data, training through a distribution network running abnormity analysis model, and acquiring corresponding training power failure number of users, training power failure duration and training complaint number;
respectively matching the number of the training power-off users, the training power-off time length and the training complaint number with a preset power-off user number table, a preset power-off time length table and a preset complaint table, and acquiring corresponding power-off user number labels, power-off time length labels and complaint number labels; the preset power outage user number table, the preset power outage duration table and the preset complaint table are respectively associated with a plurality of ordered power outage user number labels, power outage duration labels and complaint number labels, each type of label respectively corresponds to a digital range, and in addition, the ordering can be in descending order;
in addition, each status label is subjected to analytical screening, including:
matching the power failure household number label, the power failure duration label and the complaint number label with all label positions in a power failure household number table, a power failure duration table and a complaint table respectively;
if the matched power outage household number label, power outage duration label and complaint number label are positioned at the front p bits of all sequencing labels in the corresponding power outage household number table, power outage duration table and complaint table, and p is a positive integer, judging that the corresponding label is an abnormal label; otherwise, judging the label as a normal label;
constructing a distribution network operation abnormal knowledge graph by using the abnormal labels and corresponding states in the operation state data; the distribution network operation abnormal knowledge graph is constructed by the conventional means, and the specific steps are not described herein.
In the embodiment of the invention, the historical state data of a plurality of lines, the switch state data, the distribution transformer frequent fault data, the corresponding power failure household data, the power failure duration data and the complaint data are integrated and classified simultaneously by constructing the distribution network operation abnormity analysis model, so that support can be provided for the subsequent analysis and evaluation of each state data, and the abnormal state data is subjected to labeling processing and classification, so that the distribution network operation abnormity knowledge map can be constructed for the abnormal label and the corresponding state in the operation state data, and the operation rules and the equipment characteristics of the distribution networks in different areas can be mastered conveniently.
S4: calculating data of each service early warning when a distribution network operates to obtain a corresponding sensing result, wherein the sensing result comprises a distribution transformer sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate; the method comprises the following steps:
acquiring a distribution transformation total number, a switch total number, a transaction total number and a plan execution total number when the distribution network runs, and marking the distribution transformation total number, the switch total number, the transaction total number and the plan execution total number as YZi, wherein i is 1, 2, 3 and 4;
acquiring distribution transformer early warning number, switch early warning number, transaction early warning number and plan execution early warning number which occur when the distribution network operates, and marking the distribution transformer early warning number, the switch early warning number, the transaction early warning number and the plan execution early warning number as YYi; the early warning number refers to the total number of abnormal early warnings appearing in corresponding projects, and is obtained by monitoring based on the existing monitoring scheme;
carrying out simultaneous calculation on all marked data and obtaining a perception result GLi through formula calculation; the formula is:
GLi＝YYi/YZi
the sensing result comprises a distribution transformation sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate;
it should be noted that, by calculating, analyzing and evaluating the data of each service early warning, the early warning perception condition corresponding to each abnormal service can be obtained in time, so as to find the abnormality and the abnormal degree in time, the global perception rate calculation can be performed on the distribution transformer according to T +1, week, month, quarter and year, and meanwhile, the screening, viewing and deriving details can be performed according to the region selection and the perception rate threshold value, so as to facilitate viewing and adjustment.
S5: analyzing and checking the calculated sensing result in sequence to obtain a checking result and carrying out alarm prompting, wherein the method comprises the following steps:
acquiring a distribution variation constant, a switch differential constant, a differential motion constant and a plan execution differential constant of an approved distribution network during operation, and respectively combining the distribution variation constant, the switch differential constant, the differential motion constant and the plan execution differential constant with corresponding distribution total number, switch total number, differential motion total number and plan execution total number to obtain an approval result, wherein the approval result comprises a distribution variation approval rate, a switch approval rate, a differential motion approval rate and a plan execution approval rate; wherein, the calculation of the verification result is the same as the calculation mode of the perception result;
obtaining the difference value between each core fixed rate and the corresponding perception rate, and matching the difference value with the corresponding difference threshold value;
if the difference value is smaller than the difference threshold value, judging that the corresponding perception rate is effective and marking the corresponding operation type as effective perception;
if the difference value is not smaller than the difference threshold value, judging that the corresponding perception rate is invalid and marking the corresponding operation type as invalid perception;
and the difference value, the effective perception and the ineffective perception form a checking result.
In the embodiment of the invention, each monitored and evaluated perception rate and the actual approved approval rate are combined, whether the perception condition of the corresponding service is effective or not is analyzed and judged, and an alarm prompt is given in time so as to be perfect, so that the perception level of the distribution network is further improved, the research and judgment conditions of each service of the overall distribution network are obtained, and a favorable auxiliary decision is provided for the operation management and the later analysis and disposal of the fault of the distribution network;
in addition, the formulas in the embodiment of the invention are all a formula which is obtained by removing dimensions, taking the numerical value of the dimension to calculate and acquiring a large amount of data to perform software simulation to obtain the formula closest to the real situation.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential attributes thereof.
Finally, it should be noted that the above embodiments are only for illustrating the technical solutions of the present invention and not for limiting, and although the present invention is described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that modifications or equivalent substitutions may be made on the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.
Claims (10)
1. The distribution network scheduling operation management method based on big data is characterized by comprising the following steps:
acquiring data of each service when a distribution network operates and combining the data to obtain a service operation set; the service operation set comprises an OMS power failure plan, transaction management, a scheduling log, distribution network protection information, distribution network graph model topology and equipment state, fault study and judgment data, acquisition stop and power supply, distribution transformer load data and self-terminal data;
checking the operation service set to evaluate whether the plan implementation of each service is abnormal or not to obtain a service analysis set, and evaluating the implementation state of the whole service according to the service analysis set to obtain an implementation evaluation result;
training the acquired running state data through a pre-constructed distribution network running abnormity analysis model, acquiring state labels corresponding to all running states in the running state data, analyzing and screening all the state labels to obtain normal labels and abnormal labels, and constructing a distribution network running abnormity knowledge graph through the abnormal labels;
calculating data of each service early warning during the operation of the distribution network to obtain a corresponding sensing result, wherein the sensing result comprises a distribution transformer sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate;
and analyzing and checking the calculated sensing result in sequence to obtain a checking result and carrying out alarm prompt.
2. The distribution network scheduling operation management method based on big data as claimed in claim 1, wherein checking the operation service set to evaluate whether there is an abnormality in the plan implementation of each service includes:
sequentially extracting each text and number in each service item, and arranging and combining the texts and the numbers to obtain corresponding text data and digital data;
sequentially setting text names in the text data as identification marks, and respectively setting text contents corresponding to the text names and numbers in the digital data as a first verification mark and a second verification mark;
and matching the identification mark with a check list prestored in a database in sequence, and judging whether a first check mark and a second check mark associated with the identification mark are matched with sample content and sample numbers in the check list or not.
3. The distribution network scheduling operation management method based on big data according to claim 2, wherein when the matching is performed, a first matching signal is generated and a service item corresponding to the identification mark is set as a normal item; when the identification marks are not matched with the service items, generating a second matching signal and setting the service items corresponding to the identification marks as target items; the first and second matching signals, the normal item and the target item form a business analysis set.
4. The big-data-based distribution network scheduling operation management method according to claim 1, wherein the evaluation of the implementation state of the overall service according to the service analysis set comprises:
acquiring the total number of normal items and target items in a service analysis set and respectively taking value marks; matching the identification mark corresponding to the target item with a pre-constructed item name table to obtain the corresponding item weight and marking the weight; the total number of the normal items and the target items of the value marks and the item weight corresponding to the target items are simultaneously acquired to obtain the implementation degree of the whole service;
and matching the implementation degree with a preset implementation threshold value to obtain an implementation evaluation result comprising a first implementation signal, a second implementation signal and a third implementation signal.
5. The distribution network scheduling operation management method based on big data as claimed in claim 4, wherein the step of constructing the distribution network operation anomaly analysis model comprises:
acquiring a data set; the data set comprises a plurality of historical line state data, switch state data, distribution and transformation frequent fault data, corresponding power failure household data, power failure duration data and complaint data;
dividing the data set according to a preset division ratio to obtain a training set and a test set;
counting the total number N of the training sets, setting N of the training sets as a verification set, and taking the rest N-N as the training sets to be trained; n and N are positive integers, and N is more than N;
training the training set for N-N times through a neural network algorithm to obtain N-N different neural network models;
evaluating the effects of the N-N different neural network models by using the verification set, and screening out the hyper-parameters corresponding to the neural network model with the best effect;
retraining the model by using the N training sets by using the hyper-parameters with the best effect to obtain a distribution network operation abnormity analysis model; and testing and evaluating the distribution network operation abnormity analysis model by using the test set.
6. The distribution network scheduling operation management method based on big data according to claim 1, wherein the training of the acquired operation state data through a pre-constructed distribution network operation abnormity analysis model comprises:
acquiring a line state, a switch state and a distribution transformation frequent state in the running state data, training through a distribution network running abnormity analysis model, and acquiring a corresponding training power failure number of households, a training power failure duration and a training complaint number;
and respectively matching the training power failure household number, the training power failure time length and the training complaint number with a preset power failure household number table, a power failure time length table and a complaint table, and acquiring a corresponding power failure household number label, a power failure time length label and a complaint number label.
7. The distribution network scheduling operation management method based on big data according to claim 1, wherein the analyzing and screening of each status label comprises:
matching the power failure household number label, the power failure duration label and the complaint number label with all label positions in a power failure household number table, a power failure duration table and a complaint table respectively;
if the matched power outage household number label, power outage duration label and complaint number label are positioned at the front p positions of all sequencing labels in the corresponding power outage household number table, power outage duration table and complaint table, and p is a positive integer, judging that the corresponding label is an abnormal label; otherwise, judging the label as a normal label;
and constructing a distribution network operation abnormal knowledge graph by using the abnormal labels and the corresponding states in the operation state data.
8. The distribution network scheduling operation management method based on big data according to claim 6, wherein the step of calculating the data of each service early warning during the operation of the distribution network to obtain the corresponding perception result comprises the following steps:
acquiring and marking the total number of distribution transformers, the total number of switches, the total number of transaction and the total number of plan execution during the operation of the distribution network; acquiring and marking distribution transformer early warning number, switch early warning number, abnormal change early warning number and plan execution early warning number which are generated when a distribution network operates; carrying out simultaneous identification on all the marked data to obtain a sensing result; the sensing result comprises a distribution transformation sensing rate, a switch sensing rate, a transaction sensing rate and a plan execution sensing rate.
9. The distribution network scheduling operation management method based on big data as claimed in claim 1, wherein analyzing and checking the calculated sensing result to obtain a checking result and performing alarm prompt includes:
and acquiring a distribution variation constant, a switch abnormal constant, an abnormal motion abnormal constant and a plan execution abnormal constant of the checked distribution network in operation, and combining the distribution variation constant, the switch abnormal constant, the abnormal motion abnormal constant and the plan execution abnormal constant with the corresponding distribution variation total number, switch total number, abnormal motion total number and plan execution total number to obtain a checking result, wherein the checking result comprises a distribution variation checking rate, a switch checking rate, an abnormal motion checking rate and a plan execution checking rate.
10. The distribution network scheduling operation management method based on big data according to claim 9, wherein a difference between each coring rate and the corresponding sensing rate is obtained, and the difference is matched with the corresponding difference threshold;
if the difference value is smaller than the difference threshold value, judging that the corresponding perception rate is effective and marking the corresponding operation type as effective perception;
if the difference value is not smaller than the difference threshold value, judging that the corresponding perception rate is invalid and marking the corresponding operation type as invalid perception;
and the difference value, the effective perception and the ineffective perception form a checking result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210689002.0A CN114936801A (en) | 2022-06-17 | 2022-06-17 | Distribution network dispatching operation management method based on big data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210689002.0A CN114936801A (en) | 2022-06-17 | 2022-06-17 | Distribution network dispatching operation management method based on big data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114936801A true CN114936801A (en) | 2022-08-23 |
Family
ID=82868553
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210689002.0A Withdrawn CN114936801A (en) | 2022-06-17 | 2022-06-17 | Distribution network dispatching operation management method based on big data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114936801A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115577122A (en) * | 2022-11-09 | 2023-01-06 | 国网安徽省电力有限公司黄山供电公司 | Construction method of power distribution network power failure information knowledge graph |
| CN117273375A (en) * | 2023-10-19 | 2023-12-22 | 国网安徽省电力有限公司铜陵供电公司 | Distribution network fault handling decision supervision and lifting system based on knowledge graph |
| CN117391357A (en) * | 2023-10-19 | 2024-01-12 | 国网安徽省电力有限公司马鞍山供电公司 | Scheduling self-checking system for power grid scheduling operation management based on big data |
-
2022
- 2022-06-17 CN CN202210689002.0A patent/CN114936801A/en not_active Withdrawn
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115577122A (en) * | 2022-11-09 | 2023-01-06 | 国网安徽省电力有限公司黄山供电公司 | Construction method of power distribution network power failure information knowledge graph |
| CN115577122B (en) * | 2022-11-09 | 2024-04-19 | 国网安徽省电力有限公司黄山供电公司 | Construction method of power outage information knowledge graph of power distribution network |
| CN117273375A (en) * | 2023-10-19 | 2023-12-22 | 国网安徽省电力有限公司铜陵供电公司 | Distribution network fault handling decision supervision and lifting system based on knowledge graph |
| CN117391357A (en) * | 2023-10-19 | 2024-01-12 | 国网安徽省电力有限公司马鞍山供电公司 | Scheduling self-checking system for power grid scheduling operation management based on big data |
| CN117273375B (en) * | 2023-10-19 | 2024-04-02 | 国网安徽省电力有限公司铜陵供电公司 | Distribution network fault handling decision supervision and lifting system based on knowledge graph |
| CN117391357B (en) * | 2023-10-19 | 2024-04-19 | 国网安徽省电力有限公司马鞍山供电公司 | Scheduling self-checking system for power grid scheduling operation management based on big data |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114936801A (en) | Distribution network dispatching operation management method based on big data | |
| CN106154209B (en) | Electrical energy meter fault prediction technique based on decision Tree algorithms | |
| CN111104981A (en) | Hydrological prediction precision evaluation method and system based on machine learning | |
| US20050143953A1 (en) | A method and system for targeting and monitoring the energy performance of manufacturing facilities | |
| CN106570778A (en) | Big data-based data integration and line loss analysis and calculation method | |
| CN107527114A (en) | A kind of circuit taiwan area exception analysis method based on big data | |
| CN206312210U (en) | A kind of status assessing system of Distribution Network Equipment | |
| CN108898311A (en) | A kind of data quality checking method towards intelligent distribution network repairing dispatching platform | |
| US20220341996A1 (en) | Method for predicting faults in power pack of complex equipment based on a hybrid prediction model | |
| Tavakoli et al. | Human reliability analysis in maintenance team of power transmission system protection | |
| CN115700636A (en) | Equipment inspection and report generation method, device, equipment and medium based on digital twin | |
| CN115689396A (en) | Pollutant discharge control method, device, equipment and medium | |
| CN115130578A (en) | Incremental rough clustering-based online evaluation method for state of power distribution equipment | |
| CN111612149A (en) | Main network line state detection method, system and medium based on decision tree | |
| CN114493238A (en) | Power supply service risk prediction method, system, storage medium and computer equipment | |
| CN112527778B (en) | Abnormal defect elimination management system and method based on abnormal information database increment | |
| CN111330871B (en) | Quality classification method and device | |
| Bordes et al. | Partially observed competing degradation processes: modeling and inference | |
| Raza et al. | A model for analyzing performance problems and root causes in the personal software process | |
| CN115809805A (en) | Power grid multi-source data processing method based on edge calculation | |
| CN110619406A (en) | Method and device for determining business abnormity | |
| CN115333232A (en) | Online monitoring method for abnormal starting and stopping state of transformer and related device | |
| CN114493037A (en) | Automatic water balance test prediction method and system | |
| CN114155412A (en) | Deep learning model iteration method, device, equipment and storage medium | |
| Ngadiman et al. | Relationship between human errors in maintenance and overall equipment effectiveness in food industries |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20220823 |