CN110349048B - Substation multidimensional data operation interactive control platform and fault handling method - Google Patents
Substation multidimensional data operation interactive control platform and fault handling method Download PDFInfo
- Publication number
- CN110349048B CN110349048B CN201910452000.8A CN201910452000A CN110349048B CN 110349048 B CN110349048 B CN 110349048B CN 201910452000 A CN201910452000 A CN 201910452000A CN 110349048 B CN110349048 B CN 110349048B
- Authority
- CN
- China
- Prior art keywords
- current
- inspection
- equipment
- current mutation
- multidimensional data
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
- G06F18/2148—Generating training patterns; Bootstrap methods, e.g. bagging or boosting characterised by the process organisation or structure, e.g. boosting cascade
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
- 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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses a transformer substation multidimensional data operation interactive control platform and a fault handling method, when a D5000 power dispatching monitoring platform monitors that current mutation occurs, analysis is carried out on monitored information to identify current mutation causes, wherein the current mutation causes comprise system occurrence accidents, normal maintenance needs and dispatching operation mode adjustment, the identified current mutation causes are informed to a dispatching center through short messages, and corresponding handling modes are adopted according to the current mutation causes. The invention efficiently and cooperatively controls the main and auxiliary intelligent inspection equipment to realize real-time accurate judgment and control of the running state of the equipment: and monitoring, decision making and action behaviors are performed in real time, so that the power grid interactivity is realized, and the optimal control and the strong self-healing are realized.
Description
Technical Field
The invention relates to the field of electric power operation, inspection and maintenance, in particular to a fault quick linkage treatment method for multi-dimensional data of a transformer substation power station.
Background
In the existing overhaul working process of the transformer substation, the development of artificial intelligence is rapid in recent years, and the data reserves and the computing equipment are rapidly developed, so that the 'big data age' is entered. The D5000 and the auxiliary control/temperature measurement system in the power system acquire main data of the whole power grid at a high speed in real time, a large amount of power grid operation data is stored for many years, the direction of 'big data of the power system' is also generated, and a opportunity is provided for researching the current mutation reason of the power system by using a machine learning algorithm.
Disclosure of Invention
The technical problem to be solved by the invention is to provide a fault quick linkage treatment method for multidimensional data of a transformer substation power station, so as to realize quick identification of reasons and treatment measures in current abrupt change, stop misoperation and invalid operation and improve maintenance efficiency.
In order to solve the technical problems, the invention adopts the following technical scheme:
the utility model provides a transformer substation's multidimensional data operation interactive control platform, includes multidimensional data operation interactive control platform host computer, inspection robot and dispatch center, multidimensional data operation interactive control platform host computer is connected with D5000 electric power dispatch monitor platform, electric power PMS system, electric power OMS system, on-line monitoring system, auxiliary control system and inspection robot in order to gather operation information and control, when D5000 electric power dispatch monitor platform monitors the electric current mutation and takes place, multidimensional data operation interactive control platform host computer carries out analytical processing to the operation information who gathers, output electric current mutation reason according to the processing result to according to electric current mutation reason control inspection robot work.
The invention also provides a fault quick linkage treatment method based on the substation multidimensional data, which adopts the substation multidimensional data operation interaction control platform to carry out linkage treatment and comprises the following steps:
s1, when a D5000 power dispatching monitoring platform monitors that current mutation occurs, analyzing and processing the collected operation information, and identifying the cause of the current mutation, wherein the cause of the current mutation comprises system accident, normal maintenance requirement and dispatching operation mode adjustment;
s2, notifying the identified current mutation reason short message to a dispatching center, and adopting a corresponding disposal mode according to the current mutation reason.
Optionally, when the host computer of the data operation interactive control platform identifies that the current mutation occurs because of system accident, the following treatment mode is adopted:
(1) Identifying the mechanical indication position of the switch, and notifying a setting person in a short message mode;
(2) Recognizing the state of the protective pressing plate, and informing a setting person in a short message mode;
(3) Identifying an indicator lamp of the protection device, and informing a setting person in a short message mode;
the inspection robot performs the following operations:
(1) Identifying a switch mechanical position;
(2) After the switch mechanical position identification task is finished, the inspection robot performs special inspection on related equipment according to the condition that the D5000 power dispatching monitoring platform generates current mutation equipment;
(3) If a manual control request exists in the middle, the current work should be immediately suspended, the manual control is obeyed, and after the manual control is finished, the current special patrol task should be continued;
(4) And informing the patrol result to the setting personnel in a short message mode.
Optionally, when the host computer of the data operation interactive control platform recognizes that the current mutation occurs because the current mutation is needed for normal maintenance, the following treatment mode is adopted:
(1) According to the actual condition of equipment current mutation in the D5000 power dispatching monitoring platform, the multidimensional data operation interaction control platform host automatically frames an accurate inspection range during inspection, and operation information in a dispatching system is fetched in real time to determine inspection starting time;
(2) Determining important equipment inspection equipment: according to the historical temperature measurement data of the robot, determining equipment which is easy to generate heat when the current is large; determining whether a history defect device exists according to the defect record of the power PMS system; dividing the two devices into important inspection devices, wherein in a later operation and maintenance scheme, the inspection frequency of the important devices is higher than that of other devices;
(3) D5000 power dispatching monitoring platform: monitoring the equipment load in the accurate inspection range in real time, and when different limits are exceeded, automatically giving corresponding alarm and inspection schemes by the multi-dimensional data operation interaction control platform host;
(4) Inspection robot: carrying out special inspection and infrared temperature measurement once a day after the maintenance equipment is re-used;
(5) The online monitoring system monitors the environmental temperature and humidity in real time;
(6) And the auxiliary control system comprises: the video real-time monitoring device is used for monitoring the field condition, automatically adjusting the high-pressure indoor auxiliary control device and adjusting the operating environment of the device according to the collected field temperature and humidity data.
Optionally, when the operation and maintenance data operation interaction control platform host identifies that the current mutation occurs because of adjustment of the scheduling operation mode, the following treatment mode is adopted:
(1) According to the actual condition of equipment current mutation in the D5000 power dispatching monitoring platform, the multidimensional data operation interaction control platform host automatically frames an accurate inspection range during inspection, and operation information in a dispatching system is fetched in real time to determine inspection starting time;
(2) The D5000 power dispatching monitoring platform monitors the equipment load in the accurate inspection range in real time, and when the equipment load exceeds different limits, the multidimensional data operation interaction control platform host automatically gives corresponding alarm and inspection schemes;
(3) The inspection robot performs switch position determination on a remote control switch through visible light identification, and performs special inspection on equipment in an accurate inspection range at regular time;
(4) The online monitoring system monitors the environmental temperature and humidity in real time;
(5) The auxiliary control system monitors the field condition of the equipment in real time through videos, automatically adjusts the high-voltage indoor auxiliary control equipment according to the collected field temperature and humidity data, and adjusts the equipment operation environment.
Optionally, the specific inspection content of the inspection robot includes: a. infrared temperature measurement; b. main transformer oil temperature, oil level, SF6 pressure value and lightning arrester leakage current; c. switch contact temperature, ambient temperature and humidity.
Optionally, the method for analyzing and processing the collected operation information by the multidimensional data operation interaction control platform host comprises the following steps:
step (1), collecting historical data under various operation modes at intervals to form original data related to current abrupt change;
step (2), extracting characteristics from the original data, determining a mutation label, determining whether current mutation occurs by using D5000 power dispatching monitoring platform current load data, extracting various electrical quantity characteristics for various operation modes, and taking the electrical quantity characteristics as characteristic input of a subsequent XGBoost algorithm, thereby forming sample data for modeling the current mutation;
step (3), adopting XGBoost algorithm, and performing model training by using the obtained sample data;
and (4) after the XGBoost model is trained and matured, forming an electric quantity characteristic capable of reflecting the current mutation of the power grid according to the real-time running information of the power grid recorded by the D5000 power dispatching monitoring platform, and inputting the XGBoost model to judge the current mutation caused by some possible faults of the power system in real time.
Optionally, the electrical quantity characteristic reflecting the sudden change of the grid current includes: the current value of the scale is acquired only when the current suddenly changes; when the current is suddenly changed, the D5000 power dispatching monitoring platform protects the content of the action signal; d5000 power dispatching monitoring platform switch tripping signal content when current suddenly changes; d5000 power dispatching monitoring platform switch remote control operation signal content when current suddenly changes; when the current is suddenly changed, the topological diagram of the transformer substation is corresponding to whether the ring can be closed and then opened; feeding back readings by all equipment of the inspection robot when the current is suddenly changed; the system readings were monitored on-line when the current suddenly changed.
Optionally, in the training process, aiming at the characteristic that the severity of two types of errors in the current mutation prediction process is different, attention coefficients are introduced to correct the loss function of the XGBoost algorithm, so that the prediction condition of the model on an unstable sample is reduced.
Optionally, a logistic function is used to probability the model output, measure the reliability of XGBoost model output, and prevent partial mispredictions.
The invention adopts the technical scheme and has the following beneficial effects:
1. the information barriers among multiple systems are opened, the power grid information resources are fully integrated, and an operation and maintenance Internet of things data application system with resource sharing, system robustness and high application efficiency is constructed: through real-time data collection, real-time calculation, real-time synchronization, relational database, multidimensional data presentation, multidimensional data analysis engine (Mondrian), construction of cache update fully acquires data among all systems, adopts a standard protocol, uploads state information to all control layers by means of a quick communication network, forms control information and decision according to an optimal control theory and algorithm in a certain intelligent control layer, and feeds back the control information and decision to intelligent equipment to complete equipment actions.
2. The intelligent inspection equipment is controlled in a high-efficiency cooperative manner, and real-time accurate judgment, management and control of the running state of the equipment are realized: and monitoring, decision making and action behaviors are performed in real time, so that the power grid interactivity is realized, and the optimal control and the strong self-healing are realized.
3. Make up the manual periodic inspection and not enough, practice thrift manpower resources, improve fortune and examine efficiency, change production management mode: the robot system, the on-line monitoring system and the auxiliary control system are combined, so that the task amount of manual inspection is greatly reduced, the pressure of manual inspection is lightened, and the inspection frequency, the operation efficiency and the accuracy are improved.
The specific technical scheme and the beneficial effects of the invention will be described in detail in the following specific embodiments.
Drawings
The invention is further described with reference to the drawings and detailed description which follow:
FIG. 1 is a flowchart of the XGboost algorithm.
Detailed Description
Example 1
The embodiment provides a transformer substation multidimensional data operation interactive control platform, which comprises a multidimensional data operation interactive control host, a patrol robot, a D5000 power dispatching monitoring platform, a power PMS system, a power OMS system, an online monitoring system and an auxiliary control system.
The invention can realize the functions of quick positioning and automatic treatment of current causes. The multidimensional data operation interaction control platform host is used for interacting relevant information of the programming operation of the transformer substation, analyzing and processing the input information, logically judging, and outputting a current mutation reason according to a processing result to realize robot inspection and equipment part maintenance of specific programming operation.
The multidimensional data operation interactive control platform host can be connected with the monitoring system through a serial port to collect monitoring system information, such as the D5000 power dispatching monitoring platform, the power PMS system, the power OMS system and the online monitoring system.
The multidimensional data operation interactive control platform host can be connected with the dispatching center or/and the centralized control center through the modem to realize resource sharing and information exchange.
Example two
The embodiment provides a fault quick linkage treatment method based on substation multidimensional data, which comprises the following steps:
s1, when the current mutation is monitored in the D5000, analyzing accidents according to the protection action information, the switch deflection and other information in the D5000, framing the accident range, informing a dispatching and production command center by a short message, and identifying that the current mutation is caused by the system accident, wherein the disposal mode is as follows:
(1) And identifying the mechanical indication position of the switch, and informing the on-site staff and related management staff in a short message mode.
(2) And identifying the state of the protective pressing plate, and informing the on-site staff and related management staff in a short message mode. And identifying the indicator lights of the protection devices, and informing the on-site workers and related management personnel in a short message mode.
Outdoor inspection robot:
(1) The mechanical position of the switch is identified.
(2) After the switch mechanical position identification task is finished, the robot system performs special inspection on related equipment according to the D5000 current mutation equipment condition. The method comprises the following steps of (a) infrared temperature measurement, (b) visible light reading and copying of main transformer oil temperature, oil level, SF6 pressure value, lightning arrester leakage current and the like, (c) indoor rail hanging robot identification and copying of switch contact temperature, (4) corresponding historical inspection data (24 h-15 days) and (5) environment temperature and humidity reading.
(3) If there is a manual control request (the fault equipment needs to be supplemented for inspection), the current work should be immediately suspended, the manual control is obeyed, and after the manual control is finished, the current special inspection task should be continued.
(4) And informing the pilot and the attendant of the patrol result in a short message mode.
S2, when the current mutation is monitored in the D5000, the platform extracts the outage time and the outage equipment according to the outage application of the oms system, judges the risk level, identifies the normal maintenance requirement for the current mutation reason and adopts a disposal mode:
(1) According to the actual condition of the abrupt change of the current of the equipment in d5000, the traffic control platform automatically frames the accurate inspection range during the inspection [ note: the following inspection strategies are all directed to the equipment within the outlined range. And the operation information in the calling system is called in real time, and the patrol start time (namely, the total shutdown operation end time) is determined.
(2) Determining important equipment inspection equipment
And 2.1, determining equipment which is easy to generate heat when the current is high according to the historical temperature measurement data of the robot.
2.2 determining if there is a history of defective devices based on the pms defect record.
And 2.3, dividing the two devices into important inspection devices, wherein in a later operation and maintenance scheme, the inspection frequency of the important devices is higher than that of other devices.
(3) And D5000 system monitors the load of the equipment in the accurate inspection range in real time, and when the load exceeds different limits, the traffic control platform automatically gives corresponding warning and inspection schemes.
(4) Robot system
Carrying out special inspection and infrared temperature measurement once a day after the maintenance equipment is re-used; (specific inspection content: a. Infrared temperature measurement; b. Visible light reading and copying of main transformer oil temperature, oil level, SF6 pressure value, lightning arrester leakage current and the like; c. Identification and copying of switch contact temperature by indoor rail hanging robot; and corresponding historical inspection data (24 h-15 days) are retrieved).
(5) Ambient temperature and humidity reading
Monitoring (main transformer) oil chromatograph and iron core current in real time, and carrying out GIS micro-water and partial discharge; (line) arrester leakage current readings; the corresponding historical monitoring data is recalled (24 h-15 days).
(6) Auxiliary control system
Video real-time monitoring equipment field conditions; and according to the collected on-site temperature and humidity data, auxiliary control equipment such as a high-pressure indoor air conditioner, a fan, a dehumidifier and the like are automatically adjusted, and the equipment operation environment is adjusted.
S3, when the current mutation is monitored in the D5000, the platform can identify that the current mutation is caused by adjustment of a dispatching operation mode and a treatment mode according to a signal of remote control operation of a switch in the D5000 and an operation mode of ring closing and ring opening firstly:
(1) Framing a precise inspection range: according to the actual condition of the abrupt change of the current of the equipment in d5000, the traffic control platform automatically frames the accurate inspection range during the inspection period [ the inspection strategies below are all aimed at the equipment within the frame range ].
(2) D5000 system
And monitoring the equipment load in the accurate inspection range in real time, and automatically giving corresponding warning and inspection schemes by the traffic control platform when different limits are exceeded.
(3) Robot system
3.1. The robot performs switch position determination on the remote-operated switch through visible light recognition. The 35kv switch is used for identifying the switch position on site by the indoor inspection robot.
3.2 carrying out special inspection and infrared temperature measurement on equipment 1 h/8 h/24 h/48 h in the accurate inspection range. (specific inspection content: a. Infrared temperature measurement; b. Visible light reading and copying of main transformer oil temperature, oil level, SF6 pressure value, lightning arrester leakage current and the like; c. Identification and copying of switch contact temperature by an indoor track hanging robot; corresponding historical inspection data (24 h-15 days) and environmental temperature and humidity reading are called).
(4) On-line monitoring system
Monitoring (main transformer) oil chromatograph and iron core current in real time, and carrying out GIS micro-water and partial discharge; (line) arrester leakage current readings; the corresponding historical monitoring data is recalled (24 h-15 days).
(5) Auxiliary control system
Video real-time monitoring equipment field conditions; and according to the collected on-site temperature and humidity data, auxiliary control equipment such as a high-pressure indoor air conditioner, a fan, a dehumidifier and the like are automatically adjusted, and the equipment operation environment is adjusted.
According to the production and operation characteristics of a power grid enterprise, the invention analyzes the actual business of the power transformation, and the accident causes of the system are as follows: and analyzing the accident according to the protection action information, the switch deflection and other information in the D5000, and framing the accident range. And the short message informs the dispatching and production command center. Normal maintenance needs: according to the current mutation, the application of the shutdown in the oms system, the schedule prediction of each stage, the related information and the remote control operation information of the related equipment in d 5000. And (3) adjusting a scheduling operation mode: the current mutation +d5000 involves remote control operation of the switch, ring closing and ring opening operation. Abnormality of equipment defect: when the two-end systems of the current mutation and the double-circuit connecting line are operated in parallel, one current is reduced, the other current is increased, and corresponding robot inspection instructions and equipment part inspection schemes are designated.
As an implementation mode, the method for analyzing and processing the collected operation information by the multidimensional data operation interactive control platform host adopts a current mutation cause judging method based on an XGBoost algorithm so as to realize rapid distinguishing of causes and fault positions during current mutation and provide a rapid judging tool for deep overhaul and fault avoidance.
As shown with reference to fig. 1, XGBoost (ExtremeGradientBoosting) is an integrated learning framework that implements accurate classification effects through iterative computation of a set of weak classifiers. XGBoost can automatically utilize the multithread parallel computation of the CPU, a regular term is added to the realization of an algorithm, the model operation efficiency is greatly improved, the generalization capability of a model is improved, and the XGBoost has obvious advantages in the distributed application. Compared with the traditional GBDT (GradientBoostDecisionTree), the XGBoost algorithm uses Taylor second-order information, adopts a parallel/multi-core calculation mode on realization, and has faster training convergence and higher accuracy. Currently, XGBoost has been widely used in store sales prediction, event classification, and customer behavior prediction.
The data and features determine the upper limit of machine learning, and the model and algorithm only approach this upper limit, so feature engineering is very important. The data collected by the PMU in the power system is in high-dimensional time sequence, and as the scale of the power system increases, the larger the data volume is, the higher the requirement on the algorithm operation speed is, so that the feature quantity is necessary to be constructed on the original data. The feature quantity is constructed according to the following principle:
1) The characteristic quantity can well reflect the occurrence of current mutation;
2) The characteristic quantity is irrelevant to the scale of the power system;
3) The calculation of the feature quantity should ensure temporal rapidity.
First, 6 characteristics were selected according to the change of the operation state in the power system before and after the current abrupt change, and referring to experience of other researchers in the characteristic selection as shown in table 1.
According to the characteristics, firstly, the pearson correlation coefficient is utilized to analyze the correlation between the characteristics so as to judge whether information redundancy exists between the characteristics, the magnitude of the correlation coefficient represents the linear correlation degree between two variables fi and fj, and the calculation formula is shown as follows:
wherein: the value range of c is-1. Typically, if |c| > 0.5, both are considered to have a linear correlation; c=0 represents a completely nonlinear correlation; c > 0 represents a positive correlation.
And finally, reserving and deleting the corresponding features according to the calculated result in the mode to carry out modeling of model feature input data.
The method for judging the current mutation cause based on the XGBoost algorithm comprises the following steps:
step (1), using D5000, PMS, OMS, on-line monitoring device to make current, load, temperature and other data of each interval under various running modes; thereby creating a large amount of raw data related to the abrupt current change.
Step (2), extracting features from the original data, and determining mutation tags; determining whether current abrupt change occurs by using D5000 system current load data; for various operation modes, various electrical quantity characteristics are extracted and used as characteristic input of a subsequent XGBoost algorithm; thereby forming a certain amount of sample data for modeling the current abrupt change.
Step (3), adopting XGBoost algorithm and improving applicability, and performing model training by using the acquired sample data; in the training process, aiming at the characteristic that the severity of two types of errors in the current mutation prediction process is different, attention coefficients are introduced to correct the loss function of the XGBoost algorithm, so that the prediction condition of the model on an unstable sample is reduced; the logistic function is used for probability of model output, is used for measuring the reliability degree of XGBoost model output and prevents partial misprediction.
Step (4), after the XGBoost model is trained and matured, according to the real-time running information of the power grid recorded by the D5000 system, an electric quantity characteristic capable of reflecting the current mutation of the power grid is formed; and inputting the XGBoost model, so that current abrupt changes caused by some possible faults of the power system can be judged in real time.
The characteristic refers to the combined electric quantity of the equipment in the station, the current value characteristics of the main transformer, the line and the switch in the time interval when the history mutation occurs are required to be extracted, and the numerical value set of the characteristic combination is called a mutation label.
The electric quantity characteristics capable of reflecting the power grid current mutation comprise the current value characteristics of main transformers, lines and switches in the time interval when the history mutation occurs, including the current values of the main transformers, the lines and the switches in the time interval when the history is normal, and the current value characteristics of the main transformers, the lines and the switches in the time interval when the history is normal, including the current values of the main transformers, the lines and the switches in the high, middle and low states.
For location determination of a current jump, when a current jump occurs, it is determined whether a current jump occurs, in particular, with a current value change fluctuation rate of devices within a substation interval, for example: if the amplitude of the Ke Yan variable 1# main current value does not match the historical synchronization or deviates from the actual empirical amplitude, then it is known that the mutation occurred in the Ke Yan variable 1# main variable.
The current mutation cause judging method based on the XGBoost algorithm has the following advantages:
(1) As XGBoost can automatically adopt multi-thread parallel computing, the operation speed is high, and the method is suitable for processing large-scale power data.
(2) Regularization items formed by tree complexity are added in the model, so that generalization capability of the regularization items is improved, and overfitting is effectively solved;
(3) Because XGBoost is a tree structure model, various data acquired in a power system do not need to be normalized, missing values caused by some reasons can be effectively processed, and the method is suitable for current abrupt change of interval current.
While the invention has been described in terms of specific embodiments, it will be appreciated by those skilled in the art that the invention is not limited to the specific embodiments described above. Any modifications which do not depart from the functional and structural principles of the present invention are intended to be included within the scope of the appended claims.
Claims (6)
1. The method for rapidly linkage treatment of faults based on substation multidimensional data adopts a substation multidimensional data operation interactive control platform to carry out linkage treatment, the substation multidimensional data operation interactive control platform comprises a multidimensional data operation interactive control platform host, a patrol robot and a dispatching center, the multidimensional data operation interactive control platform host is connected with a D5000 power dispatching monitoring platform, a power PMS system, a power OMS system, an on-line monitoring system, an auxiliary control system and the patrol robot to collect operation information and control, when the D5000 power dispatching monitoring platform monitors that current mutation occurs, the multidimensional data operation interactive control platform host analyzes and processes the collected operation information, outputs a current mutation reason according to a processing result and controls the patrol robot to work according to the current mutation reason,
the fault quick linkage treatment method comprises the following steps:
s1, when a D5000 power dispatching monitoring platform monitors that current mutation occurs, analyzing and processing the collected operation information, and identifying the cause of the current mutation, wherein the cause of the current mutation comprises system accident, normal maintenance requirement and dispatching operation mode adjustment;
s2, notifying the identified current mutation cause short message to a dispatching center, and adopting a corresponding disposal mode according to the current mutation cause;
when the main machine of the dimensional data operation interaction control platform identifies that the current mutation occurs because of system accidents, the following treatment mode is adopted:
(1) Identifying the mechanical indication position of the switch, and notifying a setting person in a short message mode;
(2) Recognizing the state of the protective pressing plate, and informing a setting person in a short message mode;
(3) Identifying an indicator lamp of the protection device, and informing a setting person in a short message mode;
the inspection robot performs the following operations:
(1) Identifying a switch mechanical position;
(2) After the switch mechanical position identification task is finished, the inspection robot performs special inspection on related equipment according to the condition that the D5000 power dispatching monitoring platform generates current mutation equipment;
(3) If a manual control request exists in the middle, the current work should be immediately suspended, the manual control is obeyed, and after the manual control is finished, the current special patrol task should be continued;
(4) Informing the patrol result to a setting person in a short message mode;
when the main machine of the dimension data operation interaction control platform recognizes that the current mutation occurs because the current mutation is needed for normal maintenance, the following treatment mode is adopted:
(1) According to the actual condition of equipment current mutation in the D5000 power dispatching monitoring platform, the multidimensional data operation interaction control platform host automatically frames an accurate inspection range during inspection, and operation information in a dispatching system is fetched in real time to determine inspection starting time;
(2) Determining important equipment inspection equipment: according to the historical temperature measurement data of the robot, determining equipment which is easy to generate heat when the current is large; determining whether a history defect device exists according to the defect record of the power PMS system; dividing the two devices into important inspection devices, wherein in a later operation and maintenance scheme, the inspection frequency of the important devices is higher than that of other devices;
(3) D5000 power dispatching monitoring platform: monitoring the equipment load in the accurate inspection range in real time, and when different limits are exceeded, automatically giving corresponding alarm and inspection schemes by the multi-dimensional data operation interaction control platform host;
(4) Inspection robot: carrying out special inspection and infrared temperature measurement once a day after the maintenance equipment is re-used;
(5) The online monitoring system monitors the environmental temperature and humidity in real time;
(6) And the auxiliary control system comprises: the video real-time monitoring device is used for monitoring the field condition, automatically adjusting the high-pressure indoor auxiliary control device and adjusting the operating environment of the device according to the collected field temperature and humidity data;
when the operation and data operation interaction control platform host computer recognizes that the current mutation occurs because of the adjustment of the dispatching operation mode, the following treatment mode is adopted:
(1) According to the actual condition of equipment current mutation in the D5000 power dispatching monitoring platform, the multidimensional data operation interaction control platform host automatically frames an accurate inspection range during inspection, and operation information in a dispatching system is fetched in real time to determine inspection starting time;
(2) The D5000 power dispatching monitoring platform monitors the equipment load in the accurate inspection range in real time, and when the equipment load exceeds different limits, the multidimensional data operation interaction control platform host automatically gives corresponding alarm and inspection schemes;
(3) The inspection robot performs switch position determination on a remote control switch through visible light identification, and performs special inspection on equipment in an accurate inspection range at regular time;
(4) The online monitoring system monitors the environmental temperature and humidity in real time;
(5) The auxiliary control system monitors the field condition of the equipment in real time through videos, automatically adjusts the high-voltage indoor auxiliary control equipment according to the collected field temperature and humidity data, and adjusts the equipment operation environment.
2. The substation multidimensional data-based fault quick linkage handling method as recited in claim 1, wherein: the special inspection content of the inspection robot comprises: a. infrared temperature measurement; b. main transformer oil temperature, oil level, SF6 pressure value and lightning arrester leakage current; c. switch contact temperature, ambient temperature and humidity.
3. The substation multidimensional data-based fault quick linkage handling method as recited in claim 1, wherein: the method for analyzing and processing the collected operation information by the multidimensional data operation interaction control platform host comprises the following steps:
step (1), collecting historical data under various operation modes at intervals to form original data related to current abrupt change;
step (2), extracting characteristics from the original data, determining a mutation label, determining whether current mutation occurs by using D5000 power dispatching monitoring platform current load data, extracting various electrical quantity characteristics for various operation modes, and taking the electrical quantity characteristics as characteristic input of a subsequent XGBoost algorithm, thereby forming sample data for modeling the current mutation;
step (3), adopting XGBoost algorithm, and performing model training by using the obtained sample data;
and (4) after the XGBoost model is trained and matured, forming an electric quantity characteristic capable of reflecting the current mutation of the power grid according to the real-time running information of the power grid recorded by the D5000 power dispatching monitoring platform, and inputting the XGBoost model to judge the current mutation caused by some possible faults of the power system in real time.
4. The substation multidimensional data-based fault quick linkage handling method as recited in claim 3, wherein: the electrical quantity characteristics reflecting the sudden change of the power grid current include: the current value of the scale is acquired only when the current suddenly changes; when the current is suddenly changed, the D5000 power dispatching monitoring platform protects the content of the action signal; d5000 power dispatching monitoring platform switch tripping signal content when current suddenly changes; d5000 power dispatching monitoring platform switch remote control operation signal content when current suddenly changes; when the current is suddenly changed, corresponding to a topological graph of the transformer substation, determining whether to close and then open the loop; feeding back readings by all equipment of the inspection robot when the current is suddenly changed; the system readings were monitored on-line when the current suddenly changed.
5. The substation multidimensional data-based fault quick linkage handling method as recited in claim 3, wherein: in the training process, aiming at the characteristic that the severity of two types of errors in the current mutation prediction process is different, attention coefficients are introduced to correct the loss function of the XGBoost algorithm, so that the prediction condition of the model on an unstable sample is reduced.
6. The substation multidimensional data-based fault quick linkage handling method as recited in claim 3, wherein: a logistic function is used to probabilistic the model output for measuring how reliable the XGBoost model output is.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910452000.8A CN110349048B (en) | 2019-05-28 | 2019-05-28 | Substation multidimensional data operation interactive control platform and fault handling method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910452000.8A CN110349048B (en) | 2019-05-28 | 2019-05-28 | Substation multidimensional data operation interactive control platform and fault handling method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110349048A CN110349048A (en) | 2019-10-18 |
| CN110349048B true CN110349048B (en) | 2023-05-26 |
Family
ID=68174012
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910452000.8A Active CN110349048B (en) | 2019-05-28 | 2019-05-28 | Substation multidimensional data operation interactive control platform and fault handling method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110349048B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111555162B (en) * | 2020-04-06 | 2022-02-11 | 连云港市港圣开关制造有限公司 | High-efficient intelligent power distribution operation and maintenance system |
| CN116054417B (en) * | 2023-03-16 | 2023-09-19 | 山东联行智能科技有限公司 | Monitoring system and method for transformer substation |
| CN117081249A (en) * | 2023-08-17 | 2023-11-17 | 广东正力通用电气有限公司 | Automatic monitoring management platform for power supply line and line fault identification method |
| CN117034174B (en) * | 2023-09-26 | 2023-12-29 | 国网安徽省电力有限公司经济技术研究院 | Transformer substation equipment abnormality detection method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4689752A (en) * | 1983-04-13 | 1987-08-25 | Niagara Mohawk Power Corporation | System and apparatus for monitoring and control of a bulk electric power delivery system |
| CN106685704A (en) * | 2016-12-13 | 2017-05-17 | 云南电网有限责任公司电力科学研究院 | Checking method and checking device for secondary equipment failure of transformer substations |
| CN109101906A (en) * | 2018-07-27 | 2018-12-28 | 中国南方电网有限责任公司超高压输电公司贵阳局 | A kind of converting station electric power equipment infrared image exception real-time detection method and device |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2919948B2 (en) * | 1990-11-19 | 1999-07-19 | 株式会社東芝 | Electric vehicle drive motor control device |
| CN101635453A (en) * | 2009-08-20 | 2010-01-27 | 中国电力科学研究院 | Method and device for limiting secondary arc current of extra-high voltage same tower double circuit line |
| CN103117602A (en) * | 2013-03-11 | 2013-05-22 | 辽宁省电力有限公司 | Method for automatically reporting composite accidents in relay protection intelligence online running management system |
| CN104052151B (en) * | 2014-05-14 | 2017-10-20 | 国家电网公司 | Intelligent substation aided synthesis is monitored and D5000 cooler interlock platforms |
| JP6706834B2 (en) * | 2016-01-21 | 2020-06-10 | 国立大学法人東京工業大学 | Circuit breaker |
| CN108551167B (en) * | 2018-04-25 | 2020-04-17 | 浙江大学 | XGboost algorithm-based power system transient stability discrimination method |
-
2019
- 2019-05-28 CN CN201910452000.8A patent/CN110349048B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4689752A (en) * | 1983-04-13 | 1987-08-25 | Niagara Mohawk Power Corporation | System and apparatus for monitoring and control of a bulk electric power delivery system |
| CN106685704A (en) * | 2016-12-13 | 2017-05-17 | 云南电网有限责任公司电力科学研究院 | Checking method and checking device for secondary equipment failure of transformer substations |
| CN109101906A (en) * | 2018-07-27 | 2018-12-28 | 中国南方电网有限责任公司超高压输电公司贵阳局 | A kind of converting station electric power equipment infrared image exception real-time detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110349048A (en) | 2019-10-18 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110349048B (en) | Substation multidimensional data operation interactive control platform and fault handling method | |
| CN106655522B (en) | A kind of main station system suitable for electric grid secondary equipment operation management | |
| CN112788142B (en) | Intelligent edge Internet of things gateway supporting multi-sensor access | |
| CN102638100B (en) | District power network equipment abnormal alarm signal association analysis and diagnosis method | |
| CN108564254A (en) | Controller switching equipment status visualization platform based on big data | |
| CN107124291A (en) | A kind of adjusting device monitoring analysis system and method based on big data | |
| CN106570567A (en) | Main network maintenance multi-constraint multi-target evaluation expert system and optimization method | |
| CN114498934A (en) | Transformer substation monitoring system | |
| CN114936652A (en) | Electrical equipment remote monitoring control system | |
| CN113268590A (en) | Power grid equipment running state evaluation method based on equipment portrait and integrated learning | |
| CN112180865A (en) | Intelligent distribution room and equipment monitoring system | |
| CN116029699A (en) | Power information system based on data twinning and operation and maintenance method | |
| CN115378129A (en) | Power equipment running state monitoring and evaluation system and method | |
| CN113420162B (en) | Equipment operation chain state monitoring method based on knowledge graph | |
| CN113471864A (en) | Transformer substation secondary equipment field maintenance device and method | |
| CN111626542B (en) | Big data-based detection method for generator excitation system | |
| CN110245846A (en) | Current break reason method of discrimination based on XGBoost algorithm | |
| CN116468243A (en) | Power grid trend early warning method based on monitoring operation data | |
| CN116070129A (en) | Intelligent diagnosis system for hydropower centralized control accident | |
| CN116488351A (en) | Low-voltage distribution box remote monitoring method and system based on Internet of things | |
| CN116522746A (en) | Power distribution hosting method for high-energy-consumption enterprises | |
| CN106199251A (en) | A kind of distribution network failure early warning system analyzed based on adaptive modeling and method | |
| CN114626440A (en) | Monitoring signal objectification analysis method and device | |
| Zhang et al. | Research on intelligent operation and maintenance technology of primary equipment in substation | |
| CN117113157B (en) | Platform district power consumption fault detection system based on artificial intelligence |
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 |