CN115906420A - Rural power grid transformer state evaluation method and system - Google Patents
Rural power grid transformer state evaluation method and system Download PDFInfo
- Publication number
- CN115906420A CN115906420A CN202211329174.3A CN202211329174A CN115906420A CN 115906420 A CN115906420 A CN 115906420A CN 202211329174 A CN202211329174 A CN 202211329174A CN 115906420 A CN115906420 A CN 115906420A
- Authority
- CN
- China
- Prior art keywords
- transformer
- layer
- power grid
- data
- state evaluation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Supply And Distribution Of Alternating Current (AREA)
Abstract
A multi-level assessment model of transformer states is established based on a DSmT fusion algorithm, a PSO algorithm is utilized to perform feature calculation on rural power transformer data, then a multi-task-based REST integrated model and a multi-TFG-based REST data integrated model are established, a transformer state assessment model is established through the DSmT algorithm and data training is performed, and finally the trained transformer state assessment model is utilized to perform assessment. According to the rural power grid transformer state evaluation method, the state of the rural power grid transformer can be comprehensively evaluated according to historical maintenance data and a multi-level evaluation model, and timely maintenance of the rural power grid transformer is facilitated.
Description
Technical Field
The invention relates to the field of power grid transformers, in particular to a rural power grid transformer state evaluation method and system.
Background
With the rapid development of the smart power grid, the digitization of the power system is continuously promoted, the core content of the digitization is that the supervision of the whole life cycle of the rural power grid transformer from planning to operation and decommissioning is attributed to the digital power grid, so that the management of the power rural power grid transformer is more refined. The rural power grid transformer is used as an important rural power grid transformer of a rural power grid power system, and is easily subjected to comprehensive influences of external environments such as thunder and lightning and the like and internal environments such as heat, machinery and electricity. The running state of the transformer directly influences whether the power supply is normal or not and the running safety and stability of the system. If a fault occurs, serious accidents such as fire, power failure and even explosion can be caused, and the social safety and the economy are seriously influenced. Nowadays, transformers in China evolve into state maintenance, and intelligent maintenance methods become a new development trend. The state of the power transformer needs to be evaluated by establishing a complex index system, the data types are too diverse, and under the condition of lacking relevant data, the characteristic selection of the power transformer needs to be researched and the state evaluation of the rural power grid transformer needs to be carried out.
Disclosure of Invention
In order to solve the defects in the prior art, the invention provides a rural power grid transformer state evaluation method, and aims to establish a feature selection state evaluation model based on the big data of rural power grid power distribution, establish an index system for measuring and evaluating the state of rural power grid transformer equipment, and provide a multi-level rural power grid transformer evaluation model based on nuclear features.
The invention adopts the following technical scheme.
The invention provides a rural power grid transformer state evaluation method, which comprises the following steps:
performing characteristic calculation on the rural power grid power transformer data by using a PSO algorithm to obtain nuclear characteristic data serving as training data;
constructing a transformer state evaluation model based on a DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training, and obtaining a trained transformer state evaluation model;
and evaluating the rural power grid transformer to be tested by using the trained transformer state evaluation model.
Preferably, the transformer state evaluation model adopts a progressive hierarchical structure, and comprises an index layer, a defect layer, a performance layer, a component layer and an integral layer.
Preferably, the farm power transformer data comprises online data and offline data,
the online data comprises iron core grounding current, an oil color spectrum and top layer oil temperature,
the off-line data comprises infrared test data, routing inspection data, winding frequency response test data and sleeve partial discharge data.
Preferably, the data preprocessing includes a complementary difference method and an interpolation-between method.
Preferably, the REST service framework is adopted by the transformer state evaluation model.
Preferably, step 2 comprises the steps of:
step 2.1, constructing a transformer state evaluation model and preprocessing indexes of all parts in an index layer to obtain preprocessed indexes;
step 2.2, performing relative degradation degree processing on the preprocessed indexes of each part in a fault layer, calculating degradation degree vectors corresponding to the indexes, and calculating to obtain a grade membership degree vector of the fault layer by using a DSmT fusion algorithm;
step 2.3, calculating the grade membership vector of the performance layer by utilizing a DSmT fusion algorithm to the grade membership vector of the fault layer in the performance layer;
step 2.4, calculating the grade membership vector of the performance layer according to the combination weight in the component layer and a DSmT fusion algorithm to obtain the grade membership vector of the component layer;
step 2.5, in the integral layer, combining weight processing is carried out on the level membership degree vector of the component layer by using a DSmT fusion algorithm to obtain the level membership degree vector of the integral condition;
and 2.6, evaluating the state of the transformer according to the overall condition grade membership degree vector.
Preferably, the calculation formula of the relative deterioration degree processing is:
in the formula, z im The actual measured value of the ith state index; x is a radical of a fluorine atom im Relative deterioration degree of the i-th state index; z is a radical of i0 Is the initial value of the ith state index; z is a radical of i1 Is the alarm value of the ith status indicator.
Preferably, the specific expression of the DSmT fusion algorithm is:
in the formula, X 1 And X 2 Are two high-collision evidence sources processed with relative degradation degree, and m1 () and m2 () are two high-collision evidence sources X 1 And X 2 M (A) is the basic information degree of data A, D Ω And generating an over-power set for the identification frame omega through intersection operation.
Preferably, the status indicators include four statuses, namely normal, attention, abnormal and serious abnormal, and each status corresponds to a level membership vector.
The invention provides a rural power grid transformer state evaluation system on the other hand, which comprises the following modules:
the characteristic calculation module is used for performing characteristic calculation on the rural power grid power transformer data by utilizing a PSO algorithm to obtain characteristic data serving as training data;
the model building and training module is used for building a transformer state evaluation model based on a DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training, and obtaining a trained transformer state evaluation model;
and the model evaluation module is used for evaluating the rural power grid transformer to be tested by utilizing the trained transformer state evaluation model.
Preferably, the transformer state evaluation model adopts a progressive hierarchical structure, and comprises an index layer, a defect layer, a performance layer, a component layer and an integral layer.
Preferably, the farm power transformer data comprises online data and offline data,
the online data comprises iron core grounding current, an oil color spectrum and top layer oil temperature,
the off-line data comprises infrared test data, routing inspection data, winding frequency response test data and sleeve partial discharge data.
Preferably, the data preprocessing includes a complementary difference method and an interpolation-between method.
Preferably, the REST service framework is adopted by the transformer state evaluation model.
Preferably, the model construction training module is further configured to construct a transformer state evaluation model and preprocess indexes of each component in the index layer to obtain preprocessed indexes;
the system is also used for carrying out relative degradation degree processing on the preprocessed indexes of each part on the fault layer, calculating degradation degree vectors corresponding to the indexes, and calculating to obtain grade membership degree vectors of the fault layer by using a DSmT fusion algorithm;
the DSmT fusion algorithm is used for calculating the grade membership vector of the fault layer in the performance layer to obtain the grade membership vector of the performance layer;
the method is also used for calculating the grade membership vector of the performance layer according to the combination weight in the component layer and the DSmT fusion algorithm to obtain the grade membership vector of the component layer;
the method is also used for carrying out combined weight processing on the grade membership degree vectors of the component layers by utilizing a DSmT fusion algorithm in the integral layer to obtain the grade membership degree vectors of the integral condition;
and the transformer state is also evaluated according to the overall condition grade membership vector.
Preferably, the calculation formula of the relative deterioration degree process is:
in the formula, z im Actual measured value of the ith state index; x is the number of im Relative deterioration degree of the i-th state index; z is a radical of i0 Is the initial value of the ith state index; z is a radical of i1 Is the alarm value of the ith status indicator.
Preferably, the specific expression of the DSmT fusion algorithm is:
in the formula, X 1 And X 2 Are two high-collision evidence sources processed with relative degradation degree, and m1 () and m2 () are two high-collision evidence sources X 1 And X 2 M (A) is the basic information degree of data A, D Ω And generating an overpowering set for the identification frame omega through intersection operation.
Preferably, the status indicators include four statuses, namely normal, attention, abnormal and serious abnormal, and each status corresponds to a level membership vector.
Compared with the prior art, the rural power grid transformer state evaluation method has the advantages that characteristics are selected on the basis that the rural power grid transformer evaluation operation result is guaranteed to be accurate and effective, the actual maintenance current situation of the rural power grid transformer is combined, and the existing data analysis method is utilized to establish the rural power grid transformer state evaluation model. And (4) selecting features in a data preprocessing stage, and screening effective features according to a specific evaluation framework by taking the original feature set as a starting point. The rural power grid transformer is large in complexity, various in data collection forms and large in data characteristic dimension, and the characteristic selection is very important for the research on the state evaluation of the rural power grid transformer. From the aspect of electric characteristic indexes, the electric characteristic indexes can be divided into various characteristic data such as sound, vibration, light, electricity, thermal imaging, online monitoring and the like. When the data are in failure, the data are not captured by the rural power grid transformer necessarily, and data loss is caused. Under the condition of information deficiency, the establishment of the initial state evaluation index is a precondition for effectively performing state evaluation. The method analyzes the existing power transformer fault case, extracts the important characteristics of the operation state of the rural power grid transformer, and establishes the rural power grid transformer evaluation model combining data driving and subjective judgment.
Drawings
FIG. 1 is a flow chart of a rural power grid transformer state evaluation method;
FIG. 2 is a schematic diagram of a data integration system;
FIG. 3 is a schematic diagram of a hierarchical fusion evaluation process of a rural power grid transformer;
FIG. 4 is a schematic diagram of the system of the basic indexes of each layer of the evaluation model of the implementation method;
fig. 5 is a structure diagram of a rural power grid transformer state evaluation system.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. The embodiments described herein are only some embodiments of the invention, and not all embodiments. All other embodiments, which can be derived by a person skilled in the art without inventive step, are within the scope of protection of the present invention.
Fig. 1 is a flow chart of a rural power grid transformer state evaluation method.
As shown in fig. 1, the present embodiment provides a method for evaluating a state of a rural power grid transformer, including the following steps:
step 1, collecting data of a rural power grid transformer, collecting the problems of missing, multi-modal characteristics of indexes and the like based on the data, and establishing an information decision table according to a collected actual case library. And simplifying the characteristic indexes under each fault type to obtain the nuclear characteristics representing the state of the rural power grid transformer, and optimizing the multi-mode characteristics by using a PSO algorithm. And analyzing the distribution of the obtained nuclear characteristics in each system.
And performing mechanism analysis on the rural power grid transformer, and establishing a framework of components, performances, faults and indexes. And (4) counting the actual data missing condition of the transformer and researching the missing property of the information data of the rural power grid transformer. 550 actual fault cases are collected, and a fault case library is constructed.
The rural power grid power transformer data comprises online data and offline data. The data scale, data problems and sources of the large data of the rural power grid transformer are analyzed, and the collected data have the problems of serious deletion, poor quality and the like. The online data comprises iron core grounding current, oil chromatogram, top oil temperature and the like. The off-line data comprises data such as infrared test, routing inspection data, winding frequency response test, sleeve partial discharge and the like. The data are preprocessed by methods of processing, difference value compensation, interpolation separation and the like, and then model operation is carried out on the processed data.
And 2, constructing a transformer state evaluation model based on the DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training, and obtaining the trained transformer state evaluation model.
And analyzing the online data and the offline data in the REST service framework, and establishing a multi-task-based REST integration model. Index integration application of the rural power grid transformer can verify the effectiveness of the REST integration model.
Aiming at the division of three functions such as control, monitoring, relay protection and the like in the automation of a transformer substation, the state monitoring function related to the power transformer is divided into three layers, namely a spacer layer, a process layer and a transformer substation layer. The abstract research object is divided into components such as a power transformer, a sleeve, a tap switch and the like, I/O analog quantity and control commands are sent, and interface 1 is communicated with the bay level. The spacer layer mainly classifies the maintenance means of each part, for example, the body monitoring can comprise partial discharge, liquid medium and the like. And the substation layer mainly specifically monitors related index values.
Firstly, macroscopically judging description research objects, analyzing a rural power grid transformer and determining the relation among the research objects; then classifying the rural power grid transformer maintenance methods existing in various components, and extracting necessary parameters, such as maintenance time consumption, monitoring indexes, maintenance dates and the like, as accurate records of various types of actual research objects; and finally, integrating and practicing various parts of the power transformer or the requirement of the overall evaluation function on the model, and realizing complete power transformer evaluation.
The analysis process diagram when the multi-systems involved in the industrial big data integration are integrated in series, and the data privacy in the data analysis, acquisition and data transmission processes can be kept secret for the design and management of the TFG. The TFG is a graph formed by nodes and edges, and information is described by an XML language. REST is a lightweight software architecture style of the Web. The invention provides an integrated model based on REST architecture and TFG.
FIG. 2 is a schematic diagram of a data integration system.
As shown in fig. 2, an evaluation application or the like associated with the power transformer initiates a task request, which needs to include a component possibly involved, a performance evaluation, wherein the component, an index related to the performance, a path, and a database, a table path, etc. that may need to be traversed need to be labeled. Meanwhile, a data view in a task can be used by a user to perform analysis such as aggregation, extraction and the like on the data of the multi-dimensional power transformer; after the task is sent out, the accessed task queues are set in a unified mode, queue time marking is carried out on the task, and the server side continuously polls and inquires related queue tasks from the message queues and identifies the related queue tasks. This is done using an asynchronous method. Utilizing REST to complete the separation of the access interface and the database; after receiving the list, the REST background firstly analyzes the list of the XML format request to obtain a database access path, a table field and possibly related secondary model processing which needs to be carried out, for example, some indexes may relate to threshold judgment, image identification and the like of the task. All modules need to be requested in a URL mode, and all operators carry out processing according to task distribution conditions. In addition, each operator needs to request original data from a CIM mode center; the CIM center needs to complete the storage iteration of the latest monitoring data of the rural power grid transformer according to different data updating frequencies, when the CIM global mode center receives a relevant request list, the CIM global mode center carries out data resource conversion URL on the type and the research object, completes data mapping by using the mapping rule of the research object and the data, and packages the data into the request list; and after the request list is completely filled, the list is converted into a response list, and the response list is returned to applications such as power transformer evaluation, decision and the like to complete data demand response.
The invention establishes a nuclear characteristic-based transformer multi-level characteristic fusion state evaluation method, obtains subjective and objective weights of indexes by using an association rule method and an analytic hierarchy process at an index layer, calculates the objective weights according to collected actual case samples, and finally obtains an assignment operation result of combined weights. And finally, performing information fusion on each index weight by using a DSmT method to finish multi-level evaluation of the rural power grid transformer. For the condition that index data are missing in actual evaluation, the weight is corrected by combining the relevance between indexes.
The DSmT fusion algorithm is that firstly, an identification frame omega = { mu 1, mu 2,. Mu n } is established for all possible operation results of a decision study object, and then reliability distribution with an overpowering set D omega → [0,1] called m (Y) as Y is generated through intersection operation, which is also called basic information degree distribution.
Classical DSmT combination rule: aiming at the DSmT model without adding other constraint conditions under the identification framework omega.
In the formula, X 1 And X 2 Are two high-collision evidence sources processed with relative degradation degree, and m1 () and m2 () are two high-collision evidence sources X 1 And X 2 M (A) is the basic information degree of data A, D Ω Generating a hyper-representation for the recognition framework omega by intersection operationsA set of powers.
And the grade membership vector operation result of each main component and the whole transformer on each state grade can be obtained. And a reliability criterion is introduced to avoid the problem of evaluation failure caused by small difference between the grade membership numerical values, so that the final judgment on the health state of the whole component and the rural power grid transformer is completed. Assuming that the membership vector of the functional component or the rural power grid transformer as a whole with respect to each state level is L = [ L = 1 ,l 2 ,l 3 ,l 4 ]Wherein l is j And representing the membership degree of the jth state grade, and evaluating the health of the component or the rural power grid transformer as the jth state grade. Where β represents the confidence level, referenced to a general confidence level range [0.5,0.7]The present invention sets β to 0.6.
According to the difference of data development trends of index characteristics when the health state of the transformer is degraded, the numerical value type state index can be divided into positive degradation and negative degradation. Wherein, the positive degradation index shows an increasing trend when the state of the rural power grid transformer is degraded, such as the micro-water content in oil; and the value of the negative degradation index shows a descending trend when the state of the rural power grid transformer is degraded, such as insulation resistance and the like. Considering that the type of the state index of the transformer is complex and is not beneficial to the visual comparison between indexes, the invention carries out relative degradation degree processing on the state index. The value interval of the relative deterioration degree is [0,1], and the numerical value represents the deviation degree of each index from the normal state.
Wherein, the calculation formula of the relative degradation degree processing is as follows:
in the formula, z im The actual measured value of the ith state index; x is the number of im Relative deterioration degree of the i-th state index; z is a radical of i0 Is the initial value of the ith state index; z is a radical of i1 Is the alarm value of the ith status indicator. Wherein z is i1 The value of (A) is referred to the relevant rule, if only the attention value z of the state index is given in the rule ia . For the negative degradation indicator, its warning value z i1 ＝z ia 1.5; for the forward degradation indicator, its warning value z i1 ＝1.5z ia (ii) a In addition, the randomness and ambiguity of the information in the power transformer evaluation process are considered. The state indexes are four states of normal, attention, abnormity and serious abnormity, and each state corresponds to a grade membership degree vector. And calculating to obtain a correlation numerical value between the relative degradation degree of a certain state index and the cloud models in different state grades, wherein the specific calculation formula is as follows:
wherein r represents the degree of correlation between the relative deterioration degree and the gradation cloud,
ex represents a class constraint space [ C ] formed by index classification classes of rural power grid transformers min ,C max ]，Ex＝(C min +C max )/2。
En satisfies E n ′～N(E n ,H e 2 ),E n ＝(C min +C max ) Per 6, grade cloud hyper-entropy H e The constant setting is usually made in accordance with the uncertainty of the combination evaluation index, H in the present embodiment e =0.05. And completing normalization processing to obtain four state grade membership degree vectors related to normal, attention, abnormity and serious abnormity, wherein the specific expression is as follows:
in the formula, r j Indicates that the state index is relatively badDegree of association, k, between degree of differentiation and jth level cloud model j And the grade membership degree vector which represents a j level corresponding to a certain state index.
The rural power grid transformer system is complex in structure, the layered model can solve the difference among multiple state quantities, and operations are respectively carried out on all parts, performances, fault types and the like of the rural power grid transformer. The invention establishes a multilevel transformer state evaluation model based on nuclear characteristics, and has the advantages of utilizing multilevel research: by applying the progressive hierarchical structure, the complex structure of the transformer is simplified, the pertinence evaluation of different parts is shown, and the evaluation operation result is more convincing.
Fig. 3 is a schematic diagram of a rural power grid transformer hierarchy fusion evaluation process.
As shown in fig. 3, the hierarchical structure of the transformer state evaluation model is divided into five layers, namely, a reference layer, a defect layer, a performance layer, a component layer, and an overall layer.
FIG. 4 is a schematic diagram of the system of the basic indexes of each layer of the evaluation model of the implementation method.
The component layers shown in fig. 4 include a body, a sleeve, a tap changer, a cooler, and a non-electrical protection device. Different devices correspond to different performance and fault types, such as thermal faults, electrical faults, and mechanical faults. These faults are evaluated by means of different criteria, e.g. H 2 Content, total hydrocarbon content, oil level gauge, and the like.
And 2.1, preprocessing an index layer.
Because the indexes have different conditions, the invention adopts a targeted pretreatment method aiming at the characteristics of different indexes, and the following introduces the treatment methods utilized by different types of indexes: if the similar indexes are divided into a high-medium-low voltage side or ABC three phases, the most serious value is taken for evaluation; when the index does not know the specific numerical value but knows the exceeding of the standard, assigning the index as a warning value; when the index is a null value, setting the weight of the index in each fault type as 0; inputting relevant indexes of sleeve infrared and frequency response tests and vibration signals into an independent rule judging device to obtain the membership degree of each index; and for the transformer operation data index without a threshold value, evaluating field workers by using the state [0,100], and finally performing feature level fusion evaluation.
And 2.2, carrying out relative degradation degree processing on indexes of each part from the index layer to the fault layer. Calculating degradation degree vectors corresponding to the indexes; converting the vector into membership degree vectors of different grades by using a Gaussian model; and obtaining the combination weight corresponding to each index based on an analytic hierarchy process and an association rule process, performing differentiated weight fusion on each evidence by using a DSmT (Dempster-Shafer) method, and obtaining the state of each fault type based on a reliability criterion. And the first-stage evaluation obtains a judgment operation result from the index to the fault type.
And 2.3, after the operation result of the membership degree of each fault type is obtained from the fault layer to the performance layer, obtaining the grade membership degree vector of each performance index by using an improved DSmT fusion method on the basis of equal weight. The evaluation from the fault layer to the performance layer can obtain the judgment operation result from the fault type to the performance.
And 2.4, obtaining the grade membership vector of the performance layer from the performance layer to the component layer, obtaining the grade membership vector of each component by a combination weight and DSmT fusion algorithm, and obtaining the grade membership vector of each component from the performance to the transformer through comprehensive evaluation.
And 2.5, performing four-level evaluation from the components to the whole, performing combined weight processing on the grade membership vectors of all the components of the transformer, and fusing by using a DSmT algorithm to obtain the grade membership vectors of the whole state of the transformer. When the state of the rural power grid transformer is evaluated, if the difference of the state grades is not large, if a single index has a degradation condition, the judgment operation result of the grade is obtained by using the reliability criterion. The evaluation of four levels can obtain the evaluation result of the overall state of the slave transformer, help the staff to judge the running state of the rural power grid transformer, and make a maintenance scheme.
And 2.6, evaluating the state of the transformer according to the grade membership vector of the overall condition.
And inputting the training data into the transformer state evaluation model for training to obtain the trained transformer state evaluation model.
And 3, evaluating the rural power grid transformer to be tested by using the trained transformer state evaluation model.
The invention analyzes the abnormal and attention state of the insulation performance, the thermal performance and the mechanical performance of the device body when the rural power grid transformer runs. Through the model operation provided by the invention, the transformer condition is evaluated, and faults such as winding deformation, turn layer short circuit, arc discharge and the like which may occur are provided. The staff overhauls rural power grids transformer, utilizes detection means such as impedance test and chromatographic analysis to detect the transformer.
TABLE 1
Table 1 shows the results of the model operation of the method of the present invention. As shown in table 1, after the transformer state evaluation model operates, the types and abnormal degrees of each fault are determined according to four levels: normal, warning, abnormal, severe abnormal. The model operation result shows that the frequency spectrum curve of the side a of the low-voltage winding of the rural power grid transformer has larger deviation with the phases b and c in the frequency band of 150kHz, the correlation coefficient is lower, and the integral displacement condition of the transformer coil exists. By comparison with the basic spectrum, there is a large deviation of the spectral curve between 150kHz and 1200 kHz.
The overhaul result of the worker shows that the rural power grid transformer is the change of the thermal property and the insulating property caused by the deformation and the displacement of the winding. Therefore, the method provided by the invention can effectively analyze and judge the hidden trouble of the rural power grid transformer, can give out the corresponding fault occurrence probability of the rural power grid transformer aiming at each fault type, and improves the judgment accuracy rate of the overhaul personnel on the problem occurrence of the transformer.
Fig. 5 is a block diagram of a rural power grid transformer state evaluation system.
As shown in fig. 5, the embodiment further provides a rural power grid transformer state evaluation system, which includes a feature calculation module, a model construction training module, and a model evaluation module. The system can realize the evaluation of the state of the transformer by executing the processes of the steps 1 to 3 through corresponding modules.
Compared with the prior art, the rural power grid transformer state evaluation model has the advantages that the characteristics are selected on the basis of ensuring the accuracy and effectiveness of the rural power grid transformer evaluation operation result, the actual maintenance current situation of the rural power grid transformer is combined, and the existing data analysis method is utilized to establish the rural power grid transformer state evaluation model. And (4) selecting features in a data preprocessing stage, and screening effective features according to a specific evaluation framework by taking the original feature set as a starting point. Rural power grid transformer complexity is great, and data collection form is various, and data characteristic dimension is also very big, and the characteristic selection is very important to the research of rural power grid transformer state evaluation. From the view of the electric power characteristic index, the electric power characteristic index can be divided into various characteristic data such as sound, vibration, light, electricity, thermal imaging, online monitoring and the like. When the data fails, the data are not captured by the rural power grid transformer necessarily, and data loss is caused. Under the condition of information deficiency, the establishment of the initial state evaluation index is a precondition for effectively performing state evaluation. The method analyzes the existing power transformer fault case, extracts the important characteristics of the operation state of the rural power grid transformer, and establishes the rural power grid transformer evaluation model combining data driving and subjective judgment.
The present disclosure may be systems, methods, and/or computer program products. The computer program product may include a computer-readable storage medium having computer-readable program instructions embodied thereon for causing a processor to implement various aspects of the present disclosure.
The computer readable storage medium may be a tangible device that can hold and store the instructions for use by the instruction execution device. The computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, semiconductor memory device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of the computer readable storage medium would include the following: a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a Static Random Access Memory (SRAM), a portable compact disc read-only memory (CD-ROM), a Digital Versatile Disc (DVD), a memory stick, a floppy disk, a mechanical coding device, such as punch cards or in-groove projection structures having instructions stored thereon, and any suitable combination of the foregoing. Computer-readable storage media as used herein is not to be construed as transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission medium (e.g., optical pulses through a fiber optic cable), or electrical signals transmitted through electrical wires.
The computer-readable program instructions described herein may be downloaded from a computer-readable storage medium to a respective computing/processing device, or to an external computer or external storage device via a network, such as the internet, a local area network, a wide area network, and/or a wireless network. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers and/or edge servers. The network adapter card or network interface in each computing/processing device receives computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in a computer-readable storage medium in the respective computing/processing device.
The computer program instructions for carrying out operations of the present disclosure may be assembler instructions, instruction Set Architecture (ISA) instructions, machine-related instructions, microcode, firmware instructions, state setting data, or source or object code written in any combination of one or more programming languages, including an object oriented programming language such as Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The computer-readable program instructions may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider). In some embodiments, the electronic circuitry that can execute the computer-readable program instructions implements aspects of the present disclosure by utilizing the state information of the computer-readable program instructions to personalize the electronic circuitry, such as a programmable logic circuit, a Field Programmable Gate Array (FPGA), or a Programmable Logic Array (PLA).
Various aspects of the present disclosure are described herein with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the disclosure. It will be understood that each block of the flowchart illustrations and/or block diagrams, and combinations of blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer-readable program instructions.
These computer-readable program instructions may be provided to a processor of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions/acts specified in the flowchart and/or block diagram block or blocks. These computer readable program instructions may also be stored in a computer readable storage medium that can direct a computer, programmable data processing apparatus, and/or other devices to function in a particular manner, such that the computer readable medium storing the instructions comprises an article of manufacture including instructions which implement the function/act specified in the flowchart and/or block diagram block or blocks.
The computer readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer, other programmable apparatus or other devices implement the functions/acts specified in the flowchart and/or block diagram block or blocks.
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of instructions, which comprises one or more executable instructions for implementing the specified logical function(s). In some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
Finally, it should be noted that the above embodiments are only used for illustrating the technical solutions of the present invention and not for limiting the same, and although the present invention is described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: modifications and equivalents may be made to the embodiments of the invention without departing from the spirit and scope of the invention, which is to be covered by the claims.
Claims (16)
1. A rural power grid transformer state evaluation method is characterized in that:
the method comprises the following steps:
performing characteristic calculation on the rural power grid power transformer data by using a PSO algorithm to obtain nuclear characteristic data serving as training data;
constructing a transformer state evaluation model based on a DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training, and obtaining a trained transformer state evaluation model;
and evaluating the rural power grid transformer to be tested by using the trained transformer state evaluation model.
2. The rural power grid transformer state evaluation method according to claim 1, characterized in that:
wherein, the transformer state evaluation model adopts a progressive hierarchical structure comprising an index layer, a defect layer, a performance layer, a component layer and an integral layer,
an index layer for obtaining indexes of the transformer according to the characteristic data, wherein the indexes include H 2 Content and absolute yield, CO 2 Content and Absolute yield, C 2 H 4 Content, total hydrocarbon content and absolute yield, absolute yield of CO and oil level indicator, temperature, excess resistance, contact resistance, partial discharge phase waveform,
the defect layer is used for evaluating fault types and states according to index evaluation results in the index layer, the fault types comprise arc discharge, winding deformation, winding looseness, short circuit between winding layers, overheating of a body current loop, poor contact and contact faults,
a performance layer for obtaining fault attributes according to fault type and state, wherein the fault attributes comprise thermal fault, electrical fault and mechanical fault,
a component layer for evaluating different components according to fault attributes, the components including a body, a bushing, a tap changer, a cooler, and a non-electrical protection device,
and the integral layer is used for evaluating the state of the whole rural power grid transformer.
3. The rural power grid transformer state evaluation method according to claim 2, characterized in that:
the rural power grid power transformer data includes online data and offline data,
the on-line data comprises iron core grounding current, oil chromatogram and top oil temperature,
the off-line data comprises infrared test data, routing inspection data, winding frequency response test data and sleeve partial discharge data.
4. The rural power grid transformer state evaluation method according to claim 2, characterized in that:
and the REST service framework adopted by the transformer state evaluation model.
5. The rural power grid transformer state evaluation method according to claim 2, characterized in that:
the method comprises the following steps of constructing a transformer state evaluation model based on a DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training, and obtaining the trained transformer state evaluation model:
constructing a transformer state evaluation model and preprocessing indexes of each component in an index layer to obtain preprocessed indexes;
performing relative degradation degree processing on the preprocessed indexes of each part in a fault layer, calculating degradation degree vectors corresponding to the indexes, and calculating to obtain a grade membership degree vector of the fault layer by using a DSmT fusion algorithm;
utilizing a DSmT fusion algorithm to calculate the grade membership vector of the fault layer in the performance layer to obtain the grade membership vector of the performance layer;
calculating the grade membership vector of the performance layer according to the combination weight in the component layer and a DSmT fusion algorithm to obtain the grade membership vector of the component layer;
in the integral layer, combining and weighting the grade membership degree vectors of the component layer by using a DSmT fusion algorithm to obtain the grade membership degree vectors of the integral condition;
and evaluating the state of the transformer according to the overall condition grade membership vector.
6. The rural power grid transformer state evaluation method according to claim 5, characterized in that:
the calculation formula of the relative deterioration degree processing is as follows:
in the formula, z im The actual measured value of the ith state index; x is the number of im Relative deterioration degree of the i-th state index; z is a radical of i0 Is the initial value of the ith state index; z is a radical of formula i1 Is the alarm value of the ith status indicator.
7. The rural power grid transformer state evaluation method according to claim 6, characterized in that:
the specific expression of the DSmT fusion algorithm is as follows:
wherein X1 and X2 are two high-conflict evidence sources processed by relative degradation degree, m1 () and m2 () are basic information degrees of the two high-conflict evidence sources X1 and X2, m (A) is the basic information degree of data A, and D Ω And generating an overpowering set for the identification frame omega through intersection operation.
8. The rural power grid transformer state evaluation method according to claim 6, characterized in that:
the state indexes comprise normal, attention, abnormity and serious abnormity, and each state corresponds to a grade membership vector.
9. A rural power grid transformer state evaluation system which characterized in that:
the system comprises the following modules:
the characteristic calculation module is used for carrying out characteristic calculation on the rural power grid power transformer data by utilizing a PSO algorithm to obtain characteristic data serving as training data;
the model construction training module is used for constructing a transformer state evaluation model based on the DSmT fusion algorithm, inputting training data into the transformer state evaluation model for training to obtain a trained transformer state evaluation model,
and the model evaluation module is used for evaluating the rural power grid transformer to be tested by utilizing the trained transformer state evaluation model.
10. Rural power grid transformer state evaluation system of claim 9, characterized in that:
the transformer state evaluation model adopts a progressive hierarchical structure and comprises an index layer, a defect layer, a performance layer, a component layer and an integral layer.
11. Rural power grid transformer state evaluation system of claim 10, characterized in that:
the rural power grid power transformer data includes online data and offline data,
the on-line data comprises iron core grounding current, oil chromatogram and top oil temperature,
the off-line data comprises infrared test data, routing inspection data, winding frequency response test data and sleeve partial discharge data.
12. Rural power grid transformer state evaluation system of claim 10, characterized in that:
and the REST service framework adopted by the transformer state evaluation model.
13. Rural power grid transformer state evaluation system of claim 10, characterized in that:
the model construction training module is also used for constructing a transformer state evaluation model and preprocessing indexes of each component in the index layer to obtain preprocessed indexes;
the system is also used for carrying out relative degradation degree processing on the preprocessed indexes of each component on a fault layer, calculating degradation degree vectors corresponding to the indexes, and calculating grade membership degree vectors of the fault layer by using a DSmT fusion algorithm;
the DSmT fusion algorithm is used for calculating the grade membership vector of the fault layer in the performance layer to obtain the grade membership vector of the performance layer;
the method is also used for calculating the grade membership vector of the performance layer according to the combination weight in the component layer and the DSmT fusion algorithm to obtain the grade membership vector of the component layer;
the method is also used for carrying out combined weight processing on the level membership degree vector of the component layer by utilizing a DSmT fusion algorithm in the whole layer to obtain a whole condition level membership degree vector;
and the transformer state is also evaluated according to the overall condition grade membership vector.
14. Rural power grid transformer state evaluation system of claim 13, characterized in that:
the calculation formula of the relative deterioration degree processing is as follows:
in the formula, z im The actual measured value of the ith state index; x is the number of im Relative deterioration degree of the i-th state index; z is a radical of i0 Is the initial value of the i-th state index; z is a radical of i1 Is the alarm value of the ith status indicator.
15. Rural power grid transformer state evaluation system of claim 14, characterized in that:
the specific expression of the DSmT fusion algorithm is as follows:
wherein X1 and X2 are two high-conflict evidence sources processed by relative degradation degree, m1 () and m2 () are basic information degrees of the two high-conflict evidence sources X1 and X2, m (A) is a basic information degree of data A, and D Ω And generating an overpowering set for the identification frame omega through intersection operation.
16. Rural power grid transformer state evaluation system of claim 14, characterized in that:
the state indexes comprise normal, attention, abnormity and serious abnormity, and each state corresponds to a grade membership vector.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211329174.3A CN115906420A (en) | 2022-10-27 | 2022-10-27 | Rural power grid transformer state evaluation method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211329174.3A CN115906420A (en) | 2022-10-27 | 2022-10-27 | Rural power grid transformer state evaluation method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115906420A true CN115906420A (en) | 2023-04-04 |
Family
ID=86470230
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211329174.3A Pending CN115906420A (en) | 2022-10-27 | 2022-10-27 | Rural power grid transformer state evaluation method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115906420A (en) |
-
2022
- 2022-10-27 CN CN202211329174.3A patent/CN115906420A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10962999B2 (en) | Microgrid model based automated real time simulation for market based electric power system optimization | |
| CN109583520B (en) | State evaluation method of cloud model and genetic algorithm optimization support vector machine | |
| CN114138982B (en) | Knowledge graph construction method for fault diagnosis of dry-type transformer | |
| CN114723285B (en) | Power grid equipment safety evaluation prediction method | |
| CN110826228B (en) | Regional power grid operation quality limit evaluation method | |
| CN113537415A (en) | Convertor station inspection method and device based on multi-information fusion and computer equipment | |
| CN116614177B (en) | Optical fiber state multidimensional parameter monitoring system | |
| CN113239132A (en) | Online out-of-tolerance identification method for voltage transformer | |
| CN110009208A (en) | A kind of on-pole switch complete set of equipments health state evaluation method and device based on integrated intelligent algorithm | |
| CN116992399A (en) | Power equipment operation and maintenance assessment method based on power data analysis | |
| WO2020171814A1 (en) | Event selection for power grid disturbance | |
| CN117273133A (en) | Construction method of multi-source heterogeneous data knowledge graph of power distribution network | |
| JP7257455B2 (en) | Method and monitoring system for monitoring controller decision logic performance | |
| WO2019140553A1 (en) | Method and device for determining health index of power distribution system and computer storage medium | |
| CN117148048A (en) | Power distribution network fault prediction method and system based on digital twin technology | |
| CN115906420A (en) | Rural power grid transformer state evaluation method and system | |
| CN115509255A (en) | Substation patrol unmanned aerial vehicle airline risk management and control method, device, equipment and storage medium | |
| Moradzadeh et al. | Image processing-based data integrity attack detection in dynamic line rating forecasting applications | |
| Sun et al. | Power system security pattern recognition based on phase space visualization | |
| CN112001073A (en) | Reliability analysis research method of traction power supply system | |
| Morison et al. | New tools for blackout prevention | |
| Bai et al. | Abnormal Detection Scheme of Substation Equipment based on Intelligent Fusion Terminal | |
| CN117391310B (en) | Power grid equipment operation state prediction and optimization method based on digital twin technology | |
| CN113449456B (en) | Health state assessment method for power transformer under incomplete multi-mode information | |
| Yuan et al. | Research on power grid outage risk assessment and early warning model based on intelligent decision 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 |