CN110854807B - New energy relay protection intelligent checking and early warning system based on OCR technology - Google Patents
New energy relay protection intelligent checking and early warning system based on OCR technology Download PDFInfo
- Publication number
- CN110854807B CN110854807B CN201911158950.6A CN201911158950A CN110854807B CN 110854807 B CN110854807 B CN 110854807B CN 201911158950 A CN201911158950 A CN 201911158950A CN 110854807 B CN110854807 B CN 110854807B
- Authority
- CN
- China
- Prior art keywords
- fixed value
- checking
- unit
- relay protection
- early warning
- 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
- 230000004224 protection Effects 0.000 title claims abstract description 126
- 238000005516 engineering process Methods 0.000 title claims abstract description 35
- 238000000034 method Methods 0.000 claims abstract description 56
- 238000005469 granulation Methods 0.000 claims abstract description 11
- 230000003179 granulation Effects 0.000 claims abstract description 11
- 238000004458 analytical method Methods 0.000 claims description 37
- 238000012795 verification Methods 0.000 claims description 32
- 238000001514 detection method Methods 0.000 claims description 28
- 230000011218 segmentation Effects 0.000 claims description 25
- 238000012937 correction Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000035945 sensitivity Effects 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 10
- 230000003993 interaction Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 8
- 238000003066 decision tree Methods 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 238000003709 image segmentation Methods 0.000 claims description 4
- 238000013138 pruning Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 abstract description 2
- 238000012015 optical character recognition Methods 0.000 description 29
- 230000009471 action Effects 0.000 description 9
- 230000006870 function Effects 0.000 description 7
- 239000011159 matrix material Substances 0.000 description 7
- 238000007405 data analysis Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 4
- 239000002904 solvent Substances 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000005065 mining Methods 0.000 description 2
- 230000009466 transformation Effects 0.000 description 2
- 238000012800 visualization Methods 0.000 description 2
- 230000006399 behavior Effects 0.000 description 1
- 230000033228 biological regulation Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000007123 defense Effects 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000006698 induction Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000008520 organization Effects 0.000 description 1
- 238000010248 power generation Methods 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H7/00—Emergency protective circuit arrangements specially adapted for specific types of electric machines or apparatus or for sectionalised protection of cable or line systems, and effecting automatic switching in the event of an undesired change from normal working conditions
-
- 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
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02H—EMERGENCY PROTECTIVE CIRCUIT ARRANGEMENTS
- H02H1/00—Details of emergency protective circuit arrangements
- H02H1/0092—Details of emergency protective circuit arrangements concerning the data processing means, e.g. expert systems, neural networks
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- General Health & Medical Sciences (AREA)
- Human Resources & Organizations (AREA)
- Public Health (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Water Supply & Treatment (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Emergency Protection Circuit Devices (AREA)
Abstract
The invention provides a new energy relay protection intelligent checking and early warning system based on an OCR technology, which comprises a fixed value identification module, a checking parameter module and a checking fixed value module which are sequentially connected, wherein the fixed value identification module is used for carrying out information acquisition on a relay protection fixed value list based on the OCR technology to obtain a fixed value list in a standard data format; the checking parameter module is used for collecting and checking each parameter and constructing an anti-error mechanism; and the check constant value module is used for checking and early warning the constant value output by the constant value identification module. The method is based on OCR technology to carry out granulation information acquisition on the relay protection constant value list, and the accuracy of the control constant value is controlled from the source of the device, so that the problem existing in the setting of the relay protection constant value in the station is solved; and through setting up intelligent check, realize the early warning function, reduce the safety risk of relay protection definite value, guarantee new forms of energy high efficiency safety network access operation effectively, solved the problem that exists in the check of the net definite value of involving among the prior art.
Description
Technical Field
The invention relates to the technical field of relay protection of power systems, in particular to a new energy relay protection intelligent checking system based on an OCR technology; meanwhile, the invention also relates to a new energy relay protection intelligent early warning system based on the OCR technology.
Background
With the great improvement of the distributed or centralized grid-connected capacity of the new energy power supply, the complex fault characteristics and operation modes of the new energy power supply bring new challenges to relay protection. The relay protection setting value is used as a first defense line of safe and stable operation of a power grid, and under the current practical situation of 'separation of plants and networks and unified scheduling', no matter the new energy station has problems in the in-station protection setting value and the network related setting value checking, the main performance is as follows:
firstly, when a new energy protection fixed value is set, the new energy protection fixed value is often set from the perspective of protecting the safety of power generation equipment, the new energy protection fixed value is usually designed and set to be conservative, the influence of action behaviors of the new energy protection fixed value on the stable operation of a power grid is less considered, and the problems that the protection range is not clear, the selectivity and the sensitivity are not subjected to system verification and the like exist possibly.
In recent years, new energy protection misoperation and failure events occur for many times, and power-saving companies of all networks attract high attention to new energy fixed value examination. However, the existing province, city and city integrated setting calculation platform or fixed value online check system does not relate to the service range, especially relates to network fixed values, and risks of protection refusal and misoperation exist.
Thirdly, the checking of the new energy fixed value has no mature implementation scheme from the organization form or the technical system. The examination is mainly completed by manual off-line papering or electronic filing, the accuracy and the integrity of the reported fixed value cannot be guaranteed from the source of the device, the data filing, the storage and the unified management cannot be realized, the work is heavy, and the intelligent informatization level is low.
Although many protection constant value setting and checking methods have been proposed in recent years, they have been more heavily focused on the main network. Therefore, an intelligent checking and early warning system suitable for new energy relay protection is provided to solve the problems of setting and checking the new energy station internal setting value and the new energy grid related setting value, and the technical problem to be solved by technical personnel in the field is urgently needed.
Disclosure of Invention
In view of this, the invention aims to provide an intelligent checking and early warning system for relay protection of new energy based on an OCR technology, so as to realize an early warning function by accuracy of a control setting value, and solve the problems of setting and checking of an internal setting value and an associated network setting value of a new energy station.
In order to achieve the purpose, the technical scheme of the invention is realized as follows:
the utility model provides a new forms of energy relay protection intelligence is checked and early warning system based on OCR technique, includes:
the fixed value identification module is configured to perform information acquisition, detection and extraction, identification, correction and granulation analysis on the relay protection fixed value list based on an OCR technology to obtain a fixed value list in a standard data format;
the check parameter module is connected with the fixed value identification module and is configured to collect data parameters and equipment parameters output by the fixed value identification module, check the accuracy of the equipment parameters, output check results and prompt error-proof early warning for the equipment parameters which do not accord with the rules of the equipment parameter model check library; and
and the checking constant value module is connected with the checking parameter module and is configured to check and early warn the constant value output by the constant value identification module.
Further, the fixed value identification module comprises
The information acquisition unit is configured to receive a relay protection constant value list image and judge whether the image is a constant value list; if yes, entering a detection unit; if not, discarding;
the detection unit is connected with the information acquisition unit and is configured to remove the non-fixed value character image in the relay protection fixed value single image, roughly divide the relay protection fixed value single image to generate a candidate fixed value character block, finely divide the relay protection fixed value single image and extract the texture feature of the relay protection fixed value single character;
the recognition unit is connected with the detection unit and is configured to perform over-segmentation on the candidate constant value character block obtained by the detection unit and then match the candidate constant value character block with a character model to obtain an accurate constant value item; the character model is obtained through convolutional neural network training;
the correcting unit is connected with the identifying unit and is configured to perform character correction and digital correction on the fuzzy object of the relay protection fixed value single image to obtain a determined fuzzy fixed value item; the character correction method comprises data comparison and/or empirical analysis; the digital correction method is artificial intervention; and
and the granulation unit is connected with the correction unit and is configured to obtain the fixed value list of the standard data format from the correction unit, and the information collection of the fixed value list of the standard data format is carried out according to standard format templates established by different models and types of different devices.
Further, the rough segmentation method is to pre-segment the image aiming at the characteristics of stable region, stable edge and stable color of the relay protection fixed value single character, and eliminate a form frame part, a remark character description part, a background shadow part and the like in the fixed value single character to generate a candidate fixed value character block.
Further, the fine segmentation method comprises the steps of carrying out texture feature classification on the fixed value character blocks, carrying out gradient histogram and haralick feature value analysis on candidate fixed value character blocks, extracting texture features of fixed value single characters, and completing relay protection fixed value single character detection by integrating a training decision tree classifier by using an adaboost method.
Further, the pre-segmentation method is an MSER detection method, edge segmentation based on confidence degree and Mean Shift color image segmentation.
Further, the over-segmentation method comprises the steps of over-segmenting the candidate fixed-value character blocks by adopting a method of combining connected domain analysis and stroke extraction, determining the granularity and the number of the candidate fixed-value character block elements, and accelerating the candidate segmentation paths by utilizing the over-segmentation merging and binary tree pruning methods.
Further, the check parameter module comprises
A parameter unit configured to collect data parameters required for the calibration of the setting value, the data parameters including the setting value list in the standard data format output from the setting value identification module, the device parameter information, and the system impedance; the equipment parameter information comprises primary equipment information and secondary equipment information in the new energy station;
the verification unit is connected with the parameter unit and is configured to receive the data parameters collected by the parameter unit, perform accuracy verification on the equipment parameters and perform accuracy verification on the equipment parameters according to equipment parameter model verification base rules to obtain verification results; and
and the anti-error unit is connected with the verification unit and is configured to confirm and output the verification result of the verification unit and carry out anti-error early warning prompt on the equipment parameters which do not conform to the rules of the equipment parameter model verification library.
Further, the method for collecting the device parameters comprises the steps of performing data interaction with an OMS (operation management system) to obtain or input the data into a parameter unit; the method for collecting the system impedance is to perform data interaction with a province-city integrated setting calculation platform or to record the data into a parameter unit.
Further, the check and fixed value module comprises
The basic checking unit is configured to perform sensitivity and time checking on the fixed value in the fixed value list in the standard data format according to network topology analysis, fault calculation and a fixed value checking rule base of the new energy station, and if the checking meets the requirement, the logic checking unit is started; if the checking does not meet the requirements, prompting a first-level warning;
the logic checking unit is connected with the basic checking unit and is configured to selectively and logically check the fixed value in the fixed value list in the standard data format according to a fixed value checking rule base, and if the check meets the requirement, the logic checking unit enters the rule checking unit; if the check is not in accordance with the requirement, prompting a secondary warning;
the rule checking unit is connected with the logic checking unit and is configured to perform rule checking on a fixed value in a fixed value list in the standard data format according to a fixed value checking rule base, and if the fixed value meets a range in a fixed value checking rule, the fixed value enters the early warning processing unit; if the fixed value exceeds the interval range in the fixed value check rule, prompting a three-level warning;
and the early warning processing unit is connected with the rule checking unit and is configured to give early warning prompts according to checking results of the basic checking unit, the logic checking unit and the rule checking unit and the man-made judgment of checking accuracy.
Further, the fixed value checking rule base comprises a fixed value rule base, a logic rule base and a fixed value rule base, and the fixed value rule base, the logic rule base and the fixed value rule base are respectively connected with the basic checking unit, the logic checking unit and the rule checking unit.
Compared with the prior art, the invention has the following advantages:
1. the method adopts an OCR technology to carry out granulation information acquisition on the relay protection fixed value list to generate the fixed value item in a standard format, not only can obtain the truest fixed value item from a station device source, but also avoids a large amount of artificial input, prevents the problems of incompleteness and inaccuracy existing when the new energy station reports the fixed value data, controls the accuracy of the fixed value from the device source head, strengthens the fixed value management of the new energy station, and solves the problems existing in the setting of the relay protection fixed value in the station in the prior art; meanwhile, the data of the invention is subjected to unified management of filing, induction and storage, the intelligent informatization level is low, and the working intensity of operators is greatly reduced.
2. The quality reported by the new energy station equipment parameters is directly related to the accuracy of fixed value checking, the invention carries out deep analysis and comparison of historical data aiming at the new energy station equipment parameters to form a new energy equipment parameter model check library so as to carry out equipment parameter accuracy checking and anti-error early warning, and the problems that the risk of entering errors of the new energy station equipment parameters is large and the existing large amount of historical data cannot be utilized in the prior art are solved; meanwhile, the invention optimizes the parameter reporting process and further ensures the accuracy of the basic data.
3. According to the relay protection intelligent checking system, the multi-dimensional intelligent checking is arranged, the fixed value grading early warning function is realized, a user can timely find and modify the defects of fixed value calculation and value taking, the safety risk of relay protection fixed values is reduced, the high-efficiency safe network access operation of new energy is powerfully guaranteed, and the problems in the prior art of network-related fixed value checking, such as the fact that system checking is not carried out on selectivity and sensitivity, and the risks of protection rejection and misoperation exist, are solved.
4. The invention improves the informatization intelligence level of the new energy relay protection fixed value auditing work business and shortens the fixed value checking life cycle.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts. In the drawings:
FIG. 1 is a frame diagram of a new energy relay protection intelligent checking and early warning system based on an OCR technology;
fig. 2 is a frame diagram of a new energy relay protection intelligent checking and early warning system based on an OCR technology in embodiment 1 of the present invention;
FIG. 3 is a frame diagram of a verification model information base, an equipment parameter model verification base and a customized verification rule base according to embodiment 1 of the present invention;
fig. 4 is a relay protection fixed value single information acquisition flow chart of a new energy relay protection intelligent checking and early warning system based on an OCR technology in embodiment 1 of the invention;
fig. 5 is a flowchart of the working process of the detection unit and the identification unit of the new energy relay protection intelligent checking and early warning system based on the OCR technology in the embodiment 1 of the present invention;
fig. 6 is a flowchart of a checking and setting module of the new energy relay protection intelligent checking and early warning system in embodiment 1 based on the OCR technology.
Detailed Description
The technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It is to be understood that the described embodiments are merely a few embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments of the present invention without making any creative effort, shall fall within the protection scope of the present invention.
As will be appreciated by one skilled in the art, embodiments of the present invention may implement a system, apparatus, device, method or computer program. Thus, the present invention may be embodied in the form of: entirely hardware, entirely software (including firmware, resident software, micro-code, etc.), or a combination of hardware and software.
Moreover, any number of elements in the drawings are by way of example and not by way of limitation, and any naming is by way of distinction only and not by way of limitation.
In the present invention, OCR refers to a process in which an electronic device (e.g., a scanner or a digital camera) checks characters printed on paper, determines the shape thereof by detecting dark and light patterns, and then translates the shape into computer characters by a character recognition method; the method is characterized in that characters in a paper document are converted into an image file with a black-white dot matrix in an optical mode aiming at print characters, and the characters in the image are converted into a text format through recognition software for further editing and processing by word processing software.
In the present invention, the system impedance refers to a per unit value or a named value of the system impedance.
In the invention, the OMS system refers to an intelligent scheduling management system.
In the invention, the province, city and city integrated setting calculation platform refers to setting calculation software which is required by national network companies to be uniformly deployed in province, city and county, has modules such as a setting calculation power grid model, a setting coordination principle and a fixed value single-mode board, and has functions of data interaction, setting calculation and the like.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention relates to a new energy relay protection intelligent checking and early warning system based on an OCR technology, which has the main design idea that: and performing granulation information acquisition on the relay protection constant value list based on an OCR technology, and realizing an early warning function through intelligent checking.
Through the arrangement of the overall design concept, the accuracy of the control fixed value can be controlled from the source of the station device, and the fixed value management of the new energy station is strengthened; the method has the advantages that the defects of fixed value calculation and value calculation are timely discovered and modified by a user, the safety risk of protecting the fixed value is reduced, and the new energy is effectively guaranteed to efficiently and safely enter the network for operation.
Example 1
Based on the above design concept, in one specific limiting scheme of the invention, the intelligent checking and early warning system for relay protection of new energy based on the OCR technology (as shown in fig. 1) comprises a fixed value identification module, a checking parameter module and a checking fixed value module (as shown in fig. 1). In particular, the amount of the solvent to be used,
and the fixed value identification module is configured to perform information acquisition, detection and extraction, identification, correction and granulation analysis on the relay protection fixed value list based on an OCR technology to obtain the fixed value list in a standard data format. The method adopts OCR technology to carry out granulation information acquisition on the relay protection fixed value list to generate the fixed value item in a standard format, not only can obtain the truest fixed value item from the source of the station device, but also avoids a large amount of artificial input, prevents the problems of incompleteness and inaccuracy when the fixed value data is reported by a new energy station, controls the accuracy of the fixed value from the source of the device, strengthens the fixed value management of the new energy station, and solves the problems existing in the setting of the relay protection fixed value in the station in the prior art.
And the checking parameter module is connected with the fixed value identification module and is configured to collect the data parameters and the equipment parameters output by the fixed value identification module, check the accuracy of the equipment parameters, output a checking result and prompt the equipment parameters which do not accord with the rules of the equipment parameter model checking library for anti-error early warning. Because the quality reported by the new energy station equipment parameters directly relates to the accuracy of fixed value checking, the invention carries out deep analysis and comparison on historical data aiming at the new energy station equipment parameters to form a new energy equipment parameter model checking library so as to carry out equipment parameter accuracy checking and anti-error early warning, thereby solving the problems that the new energy station equipment parameters have higher risk of error entry and cannot utilize a large amount of existing historical data in the prior art; at the same time.
And the checking fixed value module is connected with the checking parameter module and is configured to check and warn the output by the fixed value identification module. According to the relay protection intelligent checking system, the early warning function is realized by setting intelligent checking, so that a user can find and modify the defects of constant value calculation and value taking in time, the safety risk of the relay protection constant value is reduced, the new energy is powerfully guaranteed to efficiently and safely enter the network for operation, and the problems in the prior art in the network-related constant value checking are solved.
In order to further improve the performance of the intelligent checking and early warning system for relay protection of new energy based on the OCR technology, in one embodiment of the present invention, the fixed value identification module includes an information obtaining unit, a detecting unit, an identification unit, a correcting unit, and a granulation unit (as shown in fig. 2). In particular, the amount of the solvent to be used,
the information acquisition unit is configured to receive a relay protection constant value list image and judge whether the image is a constant value list; if yes, entering a detection unit; if not, discarding. Specifically, in this embodiment, a relay protection fixed value single image is input: the relay protection setting value single image refers to a jpg picture format or a pdf scanning version, a paper setting value single needs to be printed from a protection device when the relay protection setting value single image is obtained, and imaging obtaining is performed on the paper setting value single through handheld equipment or equipment such as a mobile phone and a Pad.
The detection unit is connected with the information acquisition unit and is configured to remove the non-fixed value character image in the relay protection fixed value single image, roughly divide the relay protection fixed value single image to generate a candidate fixed value character block, finely divide characters, numbers, special characters and the like in the relay protection fixed value single image fixed value character block, and extract texture features of the relay protection fixed value single character. Preferably, the rough segmentation method is to pre-segment the image aiming at the characteristics of stable region, stable edge and stable color of the relay protection fixed value single character, and eliminate a form frame part, a remark character description part, a background shadow part and the like in the fixed value single character to generate a candidate fixed value character block. Preferably, the pre-segmentation method comprises an MSER detection method, edge segmentation based on confidence degree and Mean Shift color image segmentation, and three segmentation results are fused to generate the candidate constant value character block.
The fine segmentation method comprises the steps of carrying out texture feature classification on a fixed value character block, carrying out gradient histogram and haralick feature value analysis on a candidate fixed value character block, extracting texture features of fixed value single characters, and finishing relay protection fixed value single character detection by utilizing an adaboost method integrated training decision tree classifier. Specifically, in this embodiment, the specific method for integrally training the decision tree classifier by using the adaboost method is as follows:
establishing a RetinaNet basic network model framework, having a five-layer-level feature pyramid, detecting by using feature maps of different convolutional layers, then sending each layer of feature map of the feature pyramid into a regression classification sub-network for classification and regression, carrying out a positive and negative sample matching strategy, and carrying out detection and classification through a training objective function.
And the identification unit is connected with the detection unit and is configured to perform over-segmentation on the candidate constant value character block obtained by the detection unit so as to solve the speed problem that the constant value single character block is adhered and over-segmented, and then the constant value single character block is matched with the character model so as to obtain an accurate constant value item. Preferably, in this embodiment, the method of combining connected domain analysis and segment extraction is used to perform over-segmentation on the candidate fixed-value character blocks, determine the granularity and the number of the candidate fixed-value character block primitives, and accelerate the candidate segmentation paths by using the method of over-segmentation merging and binary tree pruning. More preferably, in this embodiment, the accuracy of the character model is further improved by using the character model trained by the convolutional neural network, and the specific method is as follows:
and the recognition problem of the text line image is converted into the problems of evaluation of candidate segmentation paths and optimal path search. Whether to continue the finer-scale segmentation of the image block is determined by the width of the connected component. Firstly, carrying out connected domain analysis on a text line image to obtain the connected domain characteristics of an image block. If the connected component width ω satisfies:
ω＞αHmean
the connected domain continues to be over-segmented by the method based on the stroke extraction. Where Hmean is the average height of the characters and α is an adjustable parameter. And merging the parts which are extracted and over-segmented based on the stroke segments according to the overlapping degree of the stroke segments in the horizontal direction. The degree of coincidence is measured by an empirical, constant threshold Th. The larger Th, the more difficult it is to merge segments. The number of generated over-segmentation character elements is large, and the adhered image can be segmented to the maximum extent. On the contrary, the segments are easy to be combined, and the number of over-cut character elements is small. If Th is too small, then the stuck characters that have been cut apart may be recombined. The Th is reasonably selected according to actual conditions to ensure the granularity and the number of the generated candidate character primitives, and the Th is recommended to be selected between 0.05 and 1.5.
And the correcting unit is connected with the identification unit and is configured to perform character correction and digital correction on the fuzzy object of the relay protection fixed value list to obtain a determined fuzzy fixed value item. The relay protection constant value list is basically standard characters and numbers, and the granular constant value list can be basically obtained through the algorithm. However, the phenomena of smearing, folding and the like in the photographing imaging are not excluded, so that the fuzzy object is corrected on the basis of detection and recognition. Preferably, because the protection devices of the same model print the same constant value single templates, the constant value character description can be corrected by using data comparison and empirical analysis, and if a constant value digital item is involved, the prompt warning information needs to be manually intervened to determine a constant value.
And the granulation unit is connected with the correction unit and is configured to enable the correction unit to obtain a fixed value list in a standard data format, and information collection of the fixed value list is carried out according to standard format templates established by different devices in different models and types, so that the next fixed value check and analysis are facilitated.
In order to further improve the performance of the intelligent checking and early warning system for relay protection of new energy based on the OCR technology, in another specific embodiment of the present invention, the checking parameter module includes a parameter unit, a checking unit and an anti-error unit (as shown in fig. 2). In particular, the amount of the solvent to be used,
and the parameter unit is configured to collect data parameters required by the fixed value check, the data parameters comprise the fixed value list in the standard data format output from the fixed value identification module, equipment parameter information and system impedance, the information is input into a check model information base, a complete data source is provided for the next fixed value check, and the check model information base is established in the check parameter module. The equipment parameter information comprises primary equipment information and secondary equipment information in the new energy station; specifically, in this embodiment, the device information includes device parameter information such as a main transformer, a box transformer, a grounding transformer, a power collection line, SVG, and a station transformer, and change information of a current transformer and a voltage transformer and protection configuration information. Preferably, the device parameter collection is obtained by interacting with an OMS system or is recorded into a parameter unit by a specified format; the system impedance is obtained through data interaction between the parameter unit and an integrated setting calculation platform in province, city and city or is directly recorded into the parameter unit (as shown in fig. 3). The invention optimizes the parameter reporting process and further ensures the accuracy of the basic data.
And the verification unit is connected with the parameter unit and is configured to receive the data parameters collected by the parameter unit, perform accuracy verification on the equipment parameters, and perform accuracy verification on the equipment parameters according to the equipment parameter model verification base rule to obtain a verification result. The equipment parameter model check library is established in the check parameter module, the check model information library provides basic data (as shown in figure 3) for the check model information library, equipment parameter analysis and comparison are carried out on different equipment types in a large amount of historical data according to the basic data, distribution characteristics, abnormal points, parameter predicted values and the like of various attribute values of parameters can be obtained, reasonable interval prediction of various parameters is obtained, reasonable interval prediction is carried out on input values of new parameters, and the equipment parameter model check library of new energy is formed. The equipment parameter model check library rule is specifically that if the input value of the new parameter is in the reasonable interval prediction, the new parameter is in accordance with the rule; if the new parameter entry value is not within the reasonable interval prediction, the rule is not met.
The anti-misoperation unit is connected to the verification unit and configured to confirm and output a verification result of the verification unit, specifically, in the embodiment, a data analysis result is displayed through a visualization means, and anti-misoperation early warning prompt is performed on equipment parameters which do not conform to the rules of the equipment parameter model verification library.
In order to further improve the performance of the intelligent checking and early warning system for the relay protection of the new energy based on the OCR technology, in one specific embodiment of the invention, the checking and value setting module establishes a value setting checking rule base, the value setting checking rule base comprises a value setting rule base, a logic rule base and a value setting rule base, and the value setting rule base, the logic rule base and the value setting rule base are respectively connected with the basic checking unit, the logic checking unit and the rule checking unit. Specifically, the check setting value module includes a basic check unit, a logic check unit, a rule check unit, and an early warning processing unit (as shown in fig. 2). In particular, the amount of the solvent to be used,
the basic checking unit is configured to perform sensitivity checking and time checking on the fixed value in the fixed value list in the standard data format according to the new energy station network topology analysis, the fault calculation and the fixed value principle library, and if the checking meets the requirement, the logic checking unit is started; if the check is not in accordance with the requirement, a first-level warning is prompted, and specifically in the embodiment, a red warning is prompted. And establishing a fixed value principle library aiming at primary equipment in the new energy field. The establishment of the constant value principle library is according to the regulations of relevant national standard, line standard and enterprise standard. For example:
the principle of quick-break protection setting of the grounding transformer of the wind power plant is as follows: (1) the current quick-break protection ensures that the power supply side of the grounding transformer has enough sensitivity when being in a phase short circuit in a minimum mode; (2) avoiding excitation surge current, and generally taking 7-10 times rated current; the time is typically 0 s.
The logic checking unit is connected with the basic checking unit and is configured to selectively and logically check the fixed value in the fixed value list in the standard data format according to a logic rule base, and if the check meets the requirement, the logic checking unit enters the rule checking unit; and if the check is not in accordance with the requirement, prompting a secondary warning, specifically in the embodiment, prompting an orange early warning. And aiming at logic rules among different fixed values with strong regularity, control words, trip matrixes and the like and analysis on the network frame relationship of the equipment in the new energy station, a logic rule base is established.
The rule checking unit is connected with the logic checking unit and is configured to perform rule checking on a fixed value in a fixed value list in the standard data format according to a fixed value rule base, and if the fixed value meets an interval range in a fixed value checking rule, the fixed value enters the early warning processing unit; and if the fixed value exceeds the interval range in the fixed value check rule, prompting three-level warning, specifically in the embodiment, prompting yellow early warning. Analyzing and classifying the existing check rules, historical check result data and historical fixed value data, mining data rules to obtain new fixed value item check rules including fixed value intervals, time intervals and the like, and establishing a fixed value rule base.
And the early warning processing unit is connected with the rule checking unit and is configured to give early warning prompts according to checking results of the basic checking unit, the logic checking unit and the rule checking unit and the man-made judgment of checking accuracy. And manual intervention can be performed according to different levels of fixed value warning, numerical analysis is performed on the primary early warning and the secondary early warning, the accuracy of checking is judged, and a fixed value checking report is issued.
The specific workflow of this embodiment is as follows, including the following steps:
step one, collecting relay protection constant value list information based on an OCR (optical character recognition) (as shown in fig. 4 and 5), wherein the information comprises all protection device constant value lists in the new energy station.
Firstly, inputting a relay protection fixed value single image: the relay protection setting value single image refers to a jpg picture format or a pdf scanning version, a paper setting value single needs to be printed from a protection device when the relay protection setting value single image is obtained, and imaging obtaining is performed on the paper setting value single through handheld equipment or equipment such as a mobile phone and a Pad.
Secondly, detecting and extracting a relay protection setting value list: the relay protection constant value single extraction is essentially a process for removing non-constant value character images to the maximum extent, and a coarse-to-fine extraction process is provided. The method specifically comprises the following steps:
rough segmentation: aiming at the characteristics of stable region, stable edge and stable color of a relay protection fixed value single character, the image is pre-segmented by adopting three different methods (MSER detection method, edge segmentation based on confidence coefficient and Mean Shift color image segmentation), a form frame part, a remarked character description part, a background shadow part and the like in a fixed value list are removed, and three segmentation results are fused to generate a candidate fixed value character block.
Fine segmentation: the method comprises the steps of classifying texture features aiming at characters, numbers, special characters and the like in a fixed value character block, analyzing gradient histograms and haralick feature values of candidate fixed value character blocks, extracting texture features of fixed value single characters, and completing relay protection fixed value single character detection by integrating a training decision tree classifier by using an adaboost method.
Thirdly, identifying a relay protection constant value list: considering the speed problem of adhesion and over-segmentation of a fixed-value single character block, the character block is over-segmented by adopting a method of combining connected domain analysis and stroke extraction to determine the granularity and the number of candidate character elements, the candidate segmentation path is accelerated by utilizing the over-segmentation merging and binary tree pruning method, the accuracy of a model is improved through a character model trained through a convolutional neural network, and a precisely determined value item is obtained through comparison of a segmented character and the model.
And finally, correcting a relay protection setting value list: the relay protection constant value list is basically standard characters and numbers, and the granular constant value list can be basically obtained through the algorithm. However, the phenomena of smearing, folding and the like in the photographing imaging are not excluded, so that the fuzzy object is corrected on the basis of detection and recognition. Because the protection devices of the same model print the same constant value single templates, the constant value character description can be corrected by utilizing data comparison and empirical analysis, and if a constant value digital item is involved, the prompt alarm information needs to be manually intervened to determine a constant value.
And step two, collecting equipment parameters and verifying the accuracy, and establishing an anti-error mechanism.
First, collection of device parameters: including the primary and secondary equipment information in the new energy station. The method specifically comprises the following steps: equipment parameter information such as a main transformer, a box transformer substation, a grounding transformer substation, a current collection circuit, SVG (static var generator), a station transformer substation and the like; the transformation ratio information of the current transformer and the voltage transformer; the configuration information is protected. The collection mode of the equipment parameters is that the equipment parameters can be uploaded through an OMS parameter reporting system, and the standardized interaction specification and the data interaction standard of the system and the OMS system are established; and the data can also be directly uploaded, and a standard data reporting template is established.
Secondly, checking the accuracy of the equipment parameters: according to the previously acquired equipment parameters, analyzing and comparing the equipment parameters of different equipment types in a large amount of historical data, acquiring the distribution characteristics, abnormal points, parameter predicted values and the like of various attribute values of the parameters, displaying the data analysis result by a visualization means, carrying out reasonable interval prediction on the input values of new parameters, forming a new energy equipment parameter model check library, and carrying out anti-error early warning on reported parameters.
And thirdly, performing intelligent check on the fixed value based on the multi-dimension, and performing effective evaluation and early warning on the protection fixed value (as shown in fig. 6).
First, basic checking based on fault calculation analysis: checking the sensitivity and time of a setting value according to fault calculation analysis through network topology analysis of the new energy station, and entering logic checking if the sensitivity and time of the checking are consistent; and if the check is not met, prompting red early warning, and defining the warning as primary warning.
Secondly, logic checking based on logic rules: aiming at logic rules among different fixed values with strong regularity, control words, trip matrixes and the like and analysis on a network frame relation of equipment in the new energy station, establishing a logic rule base, checking the selectivity and the logic of the fixed values, and entering rule checking if the selectivity and the logic are both in accordance after checking; if the check is not met, an orange early warning is prompted, and secondary warning is defined.
Thirdly, checking a fixed value rule based on data analysis: analyzing and classifying the existing check rules, historical check result data and historical fixed value data, mining data rules to obtain new fixed value item check rules including fixed value intervals, time intervals and the like, and if the fixed value is checked to be within an interval range, not acting; if the fixed value exceeds the range, a yellow early warning is prompted, and a three-level warning is defined.
And finally, human intervention can be performed according to different levels of fixed value warning, numerical analysis is performed on the first-level early warning and the second-level early warning, the accuracy of checking is judged, and a fixed value checking report is issued.
Example 2
The embodiment relates to a new energy relay protection intelligent checking and early warning system based on an OCR technology, which is a specific application under embodiment 1.
Step one, carrying out information acquisition on the relay protection setting value list based on an OCR technology, and specifically comprising the following steps (as shown in FIG. 4):
firstly, selecting a scanned PDF (Portable document Format) fixed value list or picture through a default path or a manually specified path on a fixed value list screening interface of a new energy relay protection intelligent checking and early warning system based on an OCR (optical character recognition) technology, and uploading the fixed value list to the system.
Secondly, calling an OCR service, and performing recognition detection on the fixed value list by using the method of full analysis and key identification bit analysis combining rough segmentation, fine segmentation, connected domain analysis and stroke extraction described in embodiment 1, so as to generate the fixed value list in a standard data format.
And thirdly, immediately displaying the fixed value list in the standard data format obtained after analysis, and quickly correcting after detection.
And after the correction is finished, storing the fixed value list in the standard data format after the correction into a warehouse to provide a data base for subsequent application.
And finally, performing granular analysis on all the corrected constant value sheets in the standard data format through the cyclic operation of the steps, and storing the constant value sheets in a warehouse.
And step two, collecting equipment parameters and carrying out accuracy verification.
First, device parameters are collected. The collected content of the equipment parameters comprises primary and secondary equipment information in the new energy station. Specifically, in this embodiment, the method includes: the system comprises equipment parameter information such as a main transformer, a box transformer, a grounding transformer, SVG (scalable vector graphics), a station transformer and the like, current transformer and voltage transformer transformation ratio information and protection configuration information. Specifically, for the collecting line parameters, the line parameter condition and the impedance value should be reported according to the actual condition.
And secondly, checking the accuracy of equipment parameters, explaining the equipment parameters in detail by using current collecting line parameters, acquiring the distribution range of impedance values under different overhead lines and cable models, automatically identifying data information outside a normal parameter range, identifying the data information as a parameter abnormal point, and performing false alarm prevention prompt.
And thirdly, carrying out intelligent check on the fixed value based on the multi-dimension, and effectively evaluating and early warning the protection fixed value. Specifically, in this embodiment, the grounding transformer protection is taken as an example to perform example analysis on the multidimensional fixed value intelligent check, and the accuracy check of the device parameters specifically includes:
first, basic checking based on the fault calculation results: firstly, judging the protection type, selecting a setting principle, and checking the sensitivity and the action time. The relay protection fixed value intelligent checking system based on the OCR has the building function of a new energy power grid model, and generates an in-plant wiring diagram based on standard SVG primitives according to the bus wiring form, the number of main transformers, the type of the main transformers and the number of the interval collection wires on each side of high, medium and low voltage sides, so that a foundation is laid for fault calculation. Specifically, in this embodiment, for the protection of the zero sequence current I section iop.i of the grounding transformer, the checking principle requires that the sensitive system is not less than 2, and the action time to.i should be longer than the longest action time of the zero sequence current II of each connecting element of the bus. During checking, checking is carried out according to the single-phase earth fault current Id of the high-voltage side of the earth transformer, if ksen = Id/Iop.I >2, the requirement on sensitivity is judged to be met, and if not, a primary warning is given; if the to.I is more than the to.II + delta t (the to.II is the longest action time of the zero sequence current II of each connecting element of the bus), judging that the action time meets the requirement, and if not, giving a first-level warning. And when the sensitivity and the action time meet the requirements, entering logic check based on the logic rule.
Secondly, logic checking based on logic rules: checking control words and a tripping matrix in the grounding variable value list, wherein the checking principle is as follows according to the grounding variable protection action logic: the corresponding protection control word of the protection fixed value non-value-taking limit value is set to be 1, the corresponding trip matrix is non-0000, and the protection trip matrixes of certain protections are the same according to the setting principle, for example, the overcurrent protection and the zero sequence protection trip matrix of the grounding transformer protection are the same. If the logical relation is met, performing fixed value rule checking based on data analysis; and if the logic relation is not met, giving a secondary warning.
Thirdly, checking a fixed value rule based on data analysis: carrying out rule analysis on historical grounding deformation set values in a checking model information base, and giving rationality prediction aiming at value intervals such as sensitivity coefficients, action time and the like, wherein if the ksen of a zero-sequence current I section belongs to [2,5], if the checking exceeds a prediction range, giving a three-level warning; similarly, the tripping matrix of the protection device with the same model has an empirical value, and if the tripping matrix is checked to be different from the empirical value, a three-level warning is given. And if the value is in the prediction range and the trip matrix is the same as the empirical value, the next step is carried out.
And finally, manually confirming and intervening according to the three constant value warnings of different levels, carrying out numerical analysis on the primary early warning and the secondary early warning, judging the accuracy of checking, and issuing a constant value checking report.
Claims (9)
1. The utility model provides a new forms of energy relay protection intelligence is checked and early warning system based on OCR technique which characterized in that: the device comprises a fixed value identification module, a check parameter module and a check fixed value module; the relay protection constant value list is obtained through the steps of obtaining, detecting and extracting, identifying, correcting and performing granulation analysis on a relay protection constant value list based on an OCR technology, and a constant value list in a standard data format is obtained; the check parameter module is connected with the fixed value identification module and is configured to collect data parameters and equipment parameters output by the fixed value identification module, check the accuracy of the equipment parameters, output check results and prompt error-proof early warning for the equipment parameters which do not accord with the rules of the equipment parameter model check library; the check constant value module is connected with the check parameter module and is configured to check and warn the constant value output by the constant value identification module; the checking and value-fixing module comprises
The basic checking unit is configured to perform sensitivity checking and time checking on the fixed value in the fixed value list in the standard data format according to network topology analysis, fault calculation and a fixed value checking rule base of the new energy station, and enter the logic checking unit if the checking meets the requirement; if the checking does not meet the requirements, prompting a first-level warning;
the logic checking unit is connected with the basic checking unit and is configured to selectively and logically check the fixed value in the fixed value list in the standard data format according to a fixed value checking rule base, and if the check meets the requirement, the logic checking unit enters the rule checking unit; if the check is not in accordance with the requirement, prompting a secondary warning;
the rule checking unit is connected with the logic checking unit and is configured to perform rule checking on a fixed value in a fixed value list in the standard data format according to a fixed value checking rule base, and if the fixed value meets a range in a fixed value checking rule, the fixed value enters the early warning processing unit; if the fixed value exceeds the interval range in the fixed value check rule, prompting a three-level warning;
and the early warning processing unit is connected with the rule checking unit and is configured to give early warning prompts according to checking results of the basic checking unit, the logic checking unit and the rule checking unit and the man-made judgment of checking accuracy.
2. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 1, wherein: the fixed value identification module comprises
The information acquisition unit is configured to receive a relay protection constant value list image and judge whether the image is a constant value list; if yes, entering a detection unit; if not, discarding;
the detection unit is connected with the information acquisition unit and is configured to remove the non-fixed value character image in the relay protection fixed value single image, roughly divide the relay protection fixed value single image to generate a candidate fixed value character block, finely divide the relay protection fixed value single image and extract the texture feature of the relay protection fixed value single character;
the recognition unit is connected with the detection unit and is configured to perform over-segmentation on the candidate constant value character block obtained by the detection unit and then match the candidate constant value character block with a character model to obtain an accurate constant value item; the character model is obtained through convolutional neural network training;
the correcting unit is connected with the identifying unit and is configured to perform character correction and digital correction on the fuzzy object of the relay protection fixed value single image to obtain a determined fuzzy fixed value item; the character correction method comprises data comparison and empirical analysis; the method of digital correction is human intervention;
and the granulation unit is connected with the correction unit and is configured to obtain the fixed value list of the standard data format from the correction unit, and the information collection of the fixed value list of the standard data format is carried out according to standard format templates established by different models and types of different devices.
3. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 2, wherein:
the rough segmentation method is characterized in that the image is pre-segmented aiming at the characteristics of stable region, stable edge and stable color of a relay protection fixed value single character, a form frame part, a remark character description part and a background shadow part in a fixed value single character are removed, and a candidate fixed value character block is generated.
4. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 3, wherein: the fine segmentation method comprises the steps of carrying out texture feature classification on a fixed value character block, carrying out gradient histogram and haralick feature value analysis on a candidate fixed value character block, extracting texture features of fixed value single characters, and finishing relay protection fixed value single character detection by utilizing an adaboost method integrated training decision tree classifier.
5. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in any one of claims 3 or 4, wherein: the pre-segmentation method comprises an MSER detection method, edge segmentation based on confidence degree and Mean Shift color image segmentation.
6. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 2, wherein: the over-segmentation method comprises the steps of over-segmenting the candidate fixed-value character blocks by adopting a method of combining connected domain analysis and stroke extraction, determining the granularity and the number of the candidate fixed-value character block elements, and accelerating the candidate segmentation paths by utilizing an over-segmentation merging and binary tree pruning method.
7. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 1, wherein: the checking parameter module comprises
A parameter unit configured to collect data parameters required for the calibration of the setting value, the data parameters including the setting value list in the standard data format output from the setting value identification module, the device parameter information, and the system impedance; the equipment parameter information comprises primary equipment information and secondary equipment information in the new energy station;
the verification unit is connected with the parameter unit and is configured to receive the data parameters collected by the parameter unit, perform accuracy verification on the equipment parameters and perform accuracy verification on the equipment parameters according to equipment parameter model verification base rules to obtain verification results;
and the anti-error unit is connected with the verification unit and is configured to confirm and output the verification result of the verification unit and carry out anti-error early warning prompt on the equipment parameters which do not conform to the rules of the equipment parameter model verification library.
8. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 7, wherein: the method for collecting the equipment parameters comprises the steps of carrying out data interaction with an OMS (operation management system) to obtain or inputting the data into a parameter unit; the method for collecting the system impedance is to perform data interaction with a province-city integrated setting calculation platform or to record the data into a parameter unit.
9. The OCR technology-based intelligent checking and early warning system for relay protection of new energy resources as claimed in claim 1, wherein: the fixed value checking rule base comprises a fixed value rule base, a logic rule base and a fixed value rule base, and the fixed value rule base, the logic rule base and the fixed value rule base are respectively connected with the basic checking unit, the logic checking unit and the rule checking unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911158950.6A CN110854807B (en) | 2019-11-22 | 2019-11-22 | New energy relay protection intelligent checking and early warning system based on OCR technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911158950.6A CN110854807B (en) | 2019-11-22 | 2019-11-22 | New energy relay protection intelligent checking and early warning system based on OCR technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110854807A CN110854807A (en) | 2020-02-28 |
| CN110854807B true CN110854807B (en) | 2021-11-12 |
Family
ID=69604023
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911158950.6A Active CN110854807B (en) | 2019-11-22 | 2019-11-22 | New energy relay protection intelligent checking and early warning system based on OCR technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110854807B (en) |
Families Citing this family (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111416438A (en) * | 2020-03-18 | 2020-07-14 | 国网湖南省电力有限公司 | Method and system for monitoring and intelligently checking state of hard pressing plate at transformer station end |
| CN111985500B (en) * | 2020-07-28 | 2024-03-29 | 国网山东省电力公司禹城市供电公司 | Verification method, system and device for relay protection fixed value input |
| CN112015720A (en) * | 2020-09-02 | 2020-12-01 | 广西电网有限责任公司河池供电局 | On-site rapid setting method and system for relay protection equipment |
| CN112255473B (en) * | 2020-09-08 | 2023-06-06 | 深圳供电局有限公司 | Relay protection device testing method and device, computer equipment and storage medium |
| CN112234587B (en) * | 2020-10-09 | 2022-07-26 | 国网河南省电力公司电力科学研究院 | Method and system for checking and evaluating relay protection setting value of power distribution network |
| CN112396052B (en) * | 2020-11-09 | 2023-05-30 | 云南电网有限责任公司昆明供电局 | Power system secondary equipment trusted fixed value list generation method and system |
| CN112436477A (en) * | 2020-11-10 | 2021-03-02 | 云南电网有限责任公司昆明供电局 | Instrument is checked fast to relay protection device definite value of transformer substation |
| CN113346449A (en) * | 2021-05-21 | 2021-09-03 | 贵州电网有限责任公司 | Protection constant value granulation intelligent checking method based on big data |
| CN113342882A (en) * | 2021-05-21 | 2021-09-03 | 贵州电网有限责任公司 | Intelligent checking system for protecting constant value granulation under big data |
| CN113346448A (en) * | 2021-05-21 | 2021-09-03 | 贵州电网有限责任公司 | Protection fixed value granulation intelligent checking system based on big data for transformer substation |
| CN113343792A (en) * | 2021-05-21 | 2021-09-03 | 贵州电网有限责任公司 | Intelligent substation protects definite value miniaturation check system |
| CN113608480A (en) * | 2021-08-11 | 2021-11-05 | 中国南方电网有限责任公司超高压输电公司贵阳局 | Fixed value checking method of extra-high voltage direct current protection device |
| CN113643154B (en) * | 2021-08-16 | 2022-10-28 | 国家电网有限公司西北分部 | Relay protection constant value intelligent auditing system |
| CN114172112A (en) * | 2021-11-16 | 2022-03-11 | 国网青海省电力公司果洛供电公司 | Relay protection constant value automatic checking and correcting device and method based on intelligent identification technology |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107276046A (en) * | 2017-06-16 | 2017-10-20 | 国网辽宁省电力有限公司电力科学研究院 | Intelligent grid realizes the abnormal Intelligent Recognition of relay protection and pre-warning system and method |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3788214B2 (en) * | 2000-08-15 | 2006-06-21 | 株式会社村田製作所 | Abnormal protection circuit for high voltage power supply for lighting discharge tube |
| CN106159911B (en) * | 2016-08-22 | 2018-07-31 | 华北电力大学 | Protect system and method in station domain based on the existing relay protective scheme amount of substation |
| CN108320128A (en) * | 2017-12-27 | 2018-07-24 | 贵州电网有限责任公司 | One kind relating to net definite value check method |
| CN108228785A (en) * | 2017-12-28 | 2018-06-29 | 中国神华能源股份有限公司 | The check method and check device of device parameter |
-
2019
- 2019-11-22 CN CN201911158950.6A patent/CN110854807B/en active Active
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107276046A (en) * | 2017-06-16 | 2017-10-20 | 国网辽宁省电力有限公司电力科学研究院 | Intelligent grid realizes the abnormal Intelligent Recognition of relay protection and pre-warning system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110854807A (en) | 2020-02-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110854807B (en) | New energy relay protection intelligent checking and early warning system based on OCR technology | |
| CN108537154B (en) | Power transmission line bird nest identification method based on HOG characteristics and machine learning | |
| CN110070530A (en) | A kind of powerline ice-covering detection method based on deep neural network | |
| CN112367273B (en) | Flow classification method and device of deep neural network model based on knowledge distillation | |
| CN116824517B (en) | Substation operation and maintenance safety control system based on visualization | |
| CN111080691A (en) | Infrared hot spot detection method and device for photovoltaic module | |
| CN105303162A (en) | Target proposed algorithm-based insulator recognition algorithm for aerial images | |
| CN115690012A (en) | Detection method for wrong connection line of electric energy meter | |
| CN114694050A (en) | Power equipment running state detection method based on infrared image | |
| CN113393442A (en) | Method and system for detecting abnormality of train parts, electronic device and storage medium | |
| CN113536944A (en) | Distribution line inspection data identification and analysis method based on image identification | |
| CN114037873A (en) | Transformer substation secondary circuit protection pressing plate detection system and method based on artificial intelligence | |
| CN113361968B (en) | Power grid infrastructure worker safety risk assessment method based on artificial intelligence and big data | |
| CN114626440A (en) | Monitoring signal objectification analysis method and device | |
| CN114495086A (en) | Method and system for identifying lightning protection monitor of electrified railway traction substation | |
| Rongrong et al. | Fault Diagnosis Method of Distribution Equipment Based on Hybrid Model of Robot and Deep Learning | |
| CN110688919A (en) | Intelligent line patrol fault identification method | |
| CN115085139B (en) | Power grid accident trip analysis method and system based on data processing | |
| CN108470140B (en) | Power transmission line bird nest identification method based on statistical characteristics and machine learning | |
| CN117095242B (en) | Intelligent building rubbish classification method and system based on machine vision | |
| Yang et al. | Wrong Wiring Detection of Electricity Meter Based on Image Processing | |
| CN116740631A (en) | Power transmission line external damage prevention detection early warning edge calculation method | |
| CN113342882A (en) | Intelligent checking system for protecting constant value granulation under big data | |
| CN116505657A (en) | Distribution line management method and system and electronic equipment | |
| CN116894822A (en) | Substation equipment defect video image preprocessing algorithm |
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 |