CN106324429B - Power distribution network fault positioning method based on RS-IA data mining - Google Patents
Power distribution network fault positioning method based on RS-IA data mining Download PDFInfo
- Publication number
- CN106324429B CN106324429B CN201610631526.9A CN201610631526A CN106324429B CN 106324429 B CN106324429 B CN 106324429B CN 201610631526 A CN201610631526 A CN 201610631526A CN 106324429 B CN106324429 B CN 106324429B
- Authority
- CN
- China
- Prior art keywords
- fault
- attribute
- decision
- distribution network
- reduction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
-
- 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
- Y04S10/52—Outage or fault management, e.g. fault detection or location
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a power distribution network fault positioning method based on RS-IA data mining, which comprises the following steps: constructing a power distribution network fault mining database; extracting fault characteristics, and determining condition attributes and decision attributes of corresponding objects, wherein an input vector set is a condition attribute set, and an output vector set forms a decision attribute set; according to the determined condition attribute and decision attribute, converting the fault mode set into an RS decision table; the problem of solving the reduction from the RS decision table is converted into the reduction of solving the minimum combination number in the distinguishing matrix; calculating the dependency degree of the decision attribute on the condition attribute; generating an initial antibody population of an immune algorithm and encoding with binary; optimal reduction obtained by using an immune algorithm; extracting fault location rules from the optimal attribute reduction set; and positioning the distribution network fault section according to the generated fault positioning rule. The invention has better fault tolerance performance.
Description
Technical Field
The invention relates to the field of power equipment fault diagnosis, in particular to a power distribution network line fault positioning method.
Background
Distribution network fault location is one of the important functions of distribution automation. After the distribution network actually breaks down, the fault location function can be used for quickly finding out the region where the fault occurs, so that effective guidance is provided for isolating the fault and recovering the power supply to the user as soon as possible, and the distribution network has important significance for improving the power supply reliability. At present, two main types of analysis methods for carrying out fault location of a power distribution network based on fault information reported by a Feeder Terminal Unit (FTU) are that one type is a unified matrix algorithm for judging and isolating fault sections of the power distribution network, and the diagnosis mode is based on sound information for fault location, is more traditional, has high reliability, and has higher requirement on memory and limited diagnosis; the other type is an artificial intelligent fault diagnosis method which is emerging in recent years, and related theories include expert systems, artificial nerves, fuzzy theories, genetic algorithms, immune algorithms and the like. The artificial intelligent fault diagnosis method can be used for diagnosing and positioning based on non-sound fault information, has stronger practicability and wider application prospect, and is technically applicable to a plurality of places needing further research.
At present, the research on the artificial intelligent fault diagnosis of the power distribution network in China is still in the primary stage, and the fault tolerance, the accuracy and the rapidity of the existing method still need to be further improved and improved. The data mining technology which is rapidly developed in recent years can provide reference information better for finding solutions for finding problem rules. Liao Zhiwei et al, discloses a "Data Mining model-based power distribution network fault location diagnosis" which applies a Data Mining (DM) model based on a combination of a Rough Set (RS) theory and a genetic algorithm to perform power distribution network fault location analysis under the condition that acquired information is lost or distorted. The comparison shows that the accuracy rate of the scheme localization diagnosis is far higher than that of a conventional artificial neural network (Artificial Neural Network, ANN) model. However, genetic algorithms tend to fall into the case of local optima, converge to local optima prematurely, and suffer from the disadvantage of "premature".
Disclosure of Invention
The invention aims to overcome the defects in the prior art and provide a power distribution network fault positioning method which is simpler, faster and more effective and has better fault tolerance. The technical proposal is as follows:
a power distribution network fault positioning method based on RS-IA data mining comprises the following steps:
1) Constructing a power distribution network fault mining database: constructing a fault mode set according to single-bit distortion information in the acquired information, and pre-constructing a single-distortion information trunk network fault positioning model with different numbers of circuit elements to serve as a decision rule database; the failure mode information is composed of: the input vector is a current out-of-limit information sequence of each switch, each switch is sequentially arranged from the incoming line switch, fault current flows through the switch before the fault element line, and '1' indicates a current out-of-limit signal; the no-current out-of-limit signal of the following switch is represented by 0; the output vector is a state sequence of the line element, and the fault state of the line is represented by '1'; the normal state is indicated by "0";
2) Extracting fault characteristics, and determining condition attributes and decision attributes of corresponding objects, wherein an input vector set is a condition attribute set, and an output vector set forms a decision attribute set;
3) According to the determined condition attribute and decision attribute, converting the fault mode set into an RS decision table;
4) The problem of solving the reduction from the RS decision table is converted into the reduction of solving the minimum combination number in the distinguishing matrix;
5) Calculating the dependency degree of the decision attribute on the condition attribute;
6) Generating an initial antibody population of an immune algorithm and encoding with binary;
7) Calculating the affinity between the antigen and the antibody, and taking the reciprocal of the affinity as an fitness function of the antibody; calculating the concentration of antibodies in the population;
8) According to the affinity and concentration, the antibody is correspondingly promoted and inhibited;
9) Storing the antibodies with high affinity and low concentration in each antibody group in a memory bank and keeping and continuously updating;
10 Performing selection, crossover and mutation operations to form a next generation parent antibody population;
11 Repeating the operations 7) to 10) until the termination condition is met, and obtaining the reduction with the minimum combination number, namely the optimal reduction obtained by using an immune algorithm;
12 Extracting fault location rules from the optimal attribute reduction set;
13 Positioning the distribution network fault section according to the generated fault positioning rule.
The invention provides a power distribution network fault positioning method based on RS-IA data mining, which is based on the existing research, and the method realizes power distribution network fault positioning under the condition of fault information variation caused by severe FTU running environment, damaged components or lost information and the like, and well solves the problem of fault positioning misjudgment caused by distortion of information acquired by a power distribution network. Compared with a genetic algorithm, the reduction performance of the immune algorithm on the rough set is better, and compared with a method based on RS-GA, the power distribution network fault positioning method based on RS-IA data mining is simpler, faster and more effective, and has better fault tolerance performance.
Drawings
Fig. 1 is a diagram of a three-power ring network open-loop operation power distribution network in an embodiment of the invention.
Figure 2 is a single trunk screen in an embodiment of the present invention.
FIG. 3 is a graph showing convergence of an immune algorithm according to an embodiment of the present invention.
Fig. 4 is a simplified flow chart of the present invention.
Detailed Description
In order to more clearly illustrate the present invention, the present invention will be further described with reference to examples and drawings.
The invention is simulated by assuming a single fault, and takes a typical 3-power ring network open-loop operation power distribution network as an example, as shown in fig. 1. In the figure, 3 circuit breakers, 2 tie switches and 16 sectionalizers are arranged, and 19 feeder lines correspond to 19 positioning sections. The circuit breaker is used as a mark, and the interconnection switch is used as a limit to divide into 3 independent power distribution areas.
And (3) carrying out comprehensive simulation on the distribution network by using a distribution fault positioning method based on RS-IA data mining. According to the construction principle of the DM fault mode, the basic fault sample mode of the line is 11 (10 line faults and wireless line fault modes), and the fault sample input vector is composed of 10 elements. Considering the principle that one bit element distortion may occur, 10 variant patterns may be derived for each basic failure mode, for a total of 110 patterns.
And global optimization search is carried out in the generalized fault mode set by using the IA, so that the optimal attribute reduction of the RS is obtained, and the specific algorithm flow is as follows. Wherein N is the size of the antibody population, m is the size of the immune memory bank, N is the length of the antibody, tac1 is the threshold value of immune selection, and MAX is the maximum iteration number, namely the termination condition for the algorithm.
1) Calculating the dependency degree K of the decision attribute D on the condition attribute C C . Order the
Sequentially removing a single attribute a E C, if K c-a ≠K c Then Core (C) =core (C)/(u) a, i.e. Core(C) A. The invention relates to a method for producing a fibre-reinforced plastic composite If K Core ＝K C Core is the best attribute reduction, otherwise step 2) is carried out. />
2) An initial population of antibodies and their encoding are generated. The invention adopts a binary coding mode, the length of the antibody, namely the number of the condition attributes C, each gene of the antibody represents the alternative state of the corresponding condition attribute, 1 represents selecting the condition attribute during reduction, and 0 represents discarding the condition attribute. When initializing, the condition attribute in the core is 1 in the corresponding bit, and the rest bits are 0 or 1 in the random way. The antibodies were expressed in [011 … 01], thereby producing an initial population of antibodies of size N.
3) The affinity was calculated. The affinity between antigen and antibody indicates how well a viable solution is satisfied for the problem, the higher the affinity, the better the solution. The affinity function selected by the invention is the reciprocal of the fitness function, and the fitness function has the following formula:
wherein n is the number of conditional attributes; l (L) v The number of '1' in the antibody, namely the number of the reduced condition attributes; alpha is a regulating factor; k is the dependence degree of the decision attribute on the condition attribute.
4) The antibody concentration was calculated. First, the affinity between two antibodies was calculated as follows:
wherein, differ vw The binding strength between two antibodies, i.e., the number of different encoded values of the genes at the same position.
The concentration of antibody V in the population is:
5) Promotion and inhibition of antibodies. To ensure diversity of antibodies, the concentration of antibodies with high affinity is increased, but too high antibodies are inhibited, and conversely the probability of production and selection of antibodies with low concentration is correspondingly increased.
6) Updating the memory bank. The high affinity, low concentration of s antibodies in each antibody population were stored in a memory pool and kept constantly updated.
7) And performing selection, crossing and mutation operations to form a next generation parent antibody group.
8) Ending when the termination condition is met, and outputting a result; otherwise turn 3).
In this example, the memory capacity is 30, the population size is 60, the crossover probability pc=0.5, the mutation probability pm=0.05, and the concentration threshold is 0.7. The maximum number of algorithm iterations is set to 150. The best fault location attributes obtained by mining are reduced to table 1.
TABLE 1 best attribute reduction for fault location correlation analysis
In order to clearly illustrate the nature of the study problem, a trunk net with 10 lines and 9 sectioning switches was extracted from the overall simulation, with one line being described as a specific example of a fault. As shown in fig. 2, where a, b … … j are lines and S1, S2 … … S9 are segment switches.
Taking the line b failure as an example, the convergence curve of the RS best reduction is obtained by using the IA method as shown in fig. 3.
The fault mode of the distribution network obtained based on the RS-IA data mining fault positioning method and the formed diagnosis result are shown in table 2. The input element is marked with distortion information bits, and the output element is marked with # is misjudged.
TABLE 2 line fault modes
As can be seen from table 2, the conventional current limit crossing signal discrimination method is very sensitive to information distortion, has no fault tolerance, and is very easy to cause erroneous discrimination and erroneous discrimination of faults. The power distribution fault positioning method based on the RS-IA data mining has higher fault tolerance capability, and can accurately realize the fault positioning of the power distribution network under most conditions.
In order to verify the superiority of the model provided by the invention, the RS-IA model is compared with a DA model method based on combination of RS and GA, a distribution fault positioning program based on the two models is compiled by MATLAB, and the trunk network shown in figure 2 is simulated to realize fault positioning. The two algorithms are basically set to be the same, the maximum iteration number of the algorithm is 150 times, and the test result is shown in table 3.
TABLE 3 Attribute reduction experiment results
As can be seen from table 3, the RS-IA model can better obtain the optimum attribute reduction and the convergence rate is faster in the same setting. And the IA algorithm can inhibit the generation of higher concentration solutions in the algorithm after introducing a concentration mechanism, prevent the algorithm from being converged to local optimum prematurely, effectively overcome the defect of 'precocity' of the genetic algorithm, and not only improve the efficiency, but also improve the accuracy. This embodiment shows that the method of the invention can realize the fault location of the distribution network in the fault information error or distortion state.
Fig. 4 is a simplified flow chart of the present invention. The following summary is made on the technical scheme of the invention:
1) And constructing a power distribution network fault mining database according to past experience and fault related information. In consideration of the fact that fault information from the FTU is easy to lose or mutate, the invention constructs a fault mode set according to the occurrence of one-bit distortion information in the acquired information, and a single-distortion information trunk network fault positioning model with unequal numbers of circuit elements is built in advance and used as a decision rule database. The failure mode information is composed of: the current out-of-limit information sequence is input for each switch (including circuit breakers, sectionalizers, tie switches, etc.). Sequentially arranging the switches from a wire inlet switch (a power supply), wherein fault current flows through the switch before a fault element line, and '1' indicates a current out-of-limit signal; the switch no current limit signal thereafter is denoted by "0". The output is a state sequence of the line element, and the fault state of the line is represented by '1'; the normal state is indicated by "0".
2) And extracting fault characteristics, and determining the condition attribute and the decision attribute of the corresponding object. Wherein the set of input vectors is a set of conditional attributes and the set of output vectors forms a set of decision attributes.
3) And converting the fault mode set into an RS decision table according to the determined condition attribute and the decision attribute.
4) The problem of reduction from the decision table is converted into reduction in which the minimum number of combinations is calculated in the discrimination matrix.
5) And calculating the dependence degree of the decision attribute D on the condition attribute C.
6) Initial antibody populations for the immunization algorithm were generated and encoded in binary.
7) Calculating the affinity between the antigen and the antibody, and taking the reciprocal of the affinity as an fitness function of the antibody; the concentration of antibody in the population was calculated.
8) Depending on the affinity and concentration, antibodies are correspondingly promoted and inhibited.
9) The high affinity and low concentration of antibodies in each antibody population are stored in a memory bank and are continuously updated.
10 Performing selection, crossover and mutation operations to form a population of next generation parent antibodies.
11 Repeating operations 7) to 10) until the termination condition is satisfied, resulting in a reduction with the smallest number of combinations, i.e., an optimal reduction using an immune algorithm.
12 Extracting fault rules from the optimal attribute reduction set.
13 Positioning the distribution network fault section according to the generated fault positioning rule.
After the distribution network fails, the FTU installed at each switch detects fault current and forms discrete fault information after comparing the fault current with a preset fault current fixed value. After the fault alarm information collected by the FTU is uploaded to the control master station, the relation between the current limit crossing information of the sectionalized switch and the position of the fault line can be analyzed according to the decision rule obtained by the RS-IA data mining model, and the line fault state corresponding to the current limit crossing information of the sectionalized switch is found out, so that the fault line of the power distribution network is correctly positioned.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the present invention.
Claims (1)
1. A power distribution network fault positioning method based on RS-IA data mining comprises the following steps:
1) Constructing a power distribution network fault mining database: constructing a fault mode set according to single-bit distortion information in the acquired information, and pre-constructing a single-distortion information trunk network fault positioning model with different numbers of circuit elements to serve as a decision rule database; the failure mode information is composed of: the input vector is a current out-of-limit information sequence of each switch, each switch is sequentially arranged from the incoming line switch, fault current flows through the switch before the fault element line, and '1' indicates a current out-of-limit signal; the no-current out-of-limit signal of the following switch is represented by 0; the output vector is a state sequence of the line element, and the fault state of the line is represented by '1'; the normal state is indicated by "0";
2) Extracting fault characteristics, and determining condition attributes and decision attributes of corresponding objects, wherein an input vector set is a condition attribute set, and an output vector set forms a decision attribute set;
3) According to the determined condition attribute and decision attribute, converting the fault mode set into an RS decision table;
4) The problem of solving the reduction from the RS decision table is converted into the reduction of solving the minimum combination number in the distinguishing matrix;
5) Calculating the dependency degree of the decision attribute on the condition attribute;
6) Generating an initial antibody population of an immune algorithm and encoding with binary;
7) Calculating the affinity between the antigen and the antibody, and taking the reciprocal of the affinity as an fitness function of the antibody; calculating the concentration of antibodies in the population;
8) According to the affinity and concentration, the antibody is correspondingly promoted and inhibited;
9) Storing the antibodies with high affinity and low concentration in each antibody group in a memory bank and keeping and continuously updating;
10 Performing selection, crossover and mutation operations to form a next generation parent antibody population;
11 Repeating the operations 7) to 10) until the termination condition is met, and obtaining the reduction with the minimum combination number, namely the optimal reduction obtained by using an immune algorithm;
12 Extracting fault location rules from the optimal attribute reduction set;
13 Positioning the distribution network fault section according to the generated fault positioning rule.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610631526.9A CN106324429B (en) | 2016-08-02 | 2016-08-02 | Power distribution network fault positioning method based on RS-IA data mining |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610631526.9A CN106324429B (en) | 2016-08-02 | 2016-08-02 | Power distribution network fault positioning method based on RS-IA data mining |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106324429A CN106324429A (en) | 2017-01-11 |
| CN106324429B true CN106324429B (en) | 2023-06-02 |
Family
ID=57740710
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610631526.9A Active CN106324429B (en) | 2016-08-02 | 2016-08-02 | Power distribution network fault positioning method based on RS-IA data mining |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106324429B (en) |
Families Citing this family (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106841927B (en) * | 2017-03-17 | 2019-05-31 | 国网江苏省电力公司宿迁供电公司 | Fault Locating Method containing distributed power distribution network |
| CN108693771A (en) * | 2017-04-10 | 2018-10-23 | 南京理工大学 | A kind of distribution network failure section location algorithm based on Multiple-population Genetic Algorithm |
| CN107579844A (en) * | 2017-08-18 | 2018-01-12 | 北京航空航天大学 | It is a kind of that failure method for digging is dynamically associated based on service path and frequency matrix |
| CN107861026B (en) * | 2017-11-02 | 2019-11-08 | 湖北工业大学 | A kind of electrical power distribution network fault location method based on hybrid artificial immune system |
| CN108649575B (en) * | 2018-06-19 | 2021-01-08 | 清华大学 | Alternating current-direct current hybrid micro-grid and protection control center and protection control method thereof |
| US11616390B2 (en) | 2018-06-19 | 2023-03-28 | Tsinghua University | Micro-grid reconstruction method and device, micro-grid protection control center, and storage medium |
| CN109490704A (en) * | 2018-10-16 | 2019-03-19 | 河海大学 | A kind of Fault Section Location of Distribution Network based on random forests algorithm |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488802A (en) * | 2013-10-16 | 2014-01-01 | 国家电网公司 | EHV (Extra-High Voltage) power grid fault rule mining method based on rough set association rule |
| CN104598968A (en) * | 2014-10-13 | 2015-05-06 | 芜湖扬宇机电技术开发有限公司 | Fault diagnosis method of transformer |
| CN104731966A (en) * | 2015-04-07 | 2015-06-24 | 河海大学 | Subway fault diagnosis method based on data mining |
| CN104931857A (en) * | 2015-06-25 | 2015-09-23 | 山东大学 | Power distribution network fault locating method based on D-S evidence theory |
-
2016
- 2016-08-02 CN CN201610631526.9A patent/CN106324429B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103488802A (en) * | 2013-10-16 | 2014-01-01 | 国家电网公司 | EHV (Extra-High Voltage) power grid fault rule mining method based on rough set association rule |
| CN104598968A (en) * | 2014-10-13 | 2015-05-06 | 芜湖扬宇机电技术开发有限公司 | Fault diagnosis method of transformer |
| CN104731966A (en) * | 2015-04-07 | 2015-06-24 | 河海大学 | Subway fault diagnosis method based on data mining |
| CN104931857A (en) * | 2015-06-25 | 2015-09-23 | 山东大学 | Power distribution network fault locating method based on D-S evidence theory |
Non-Patent Citations (2)
| Title |
|---|
| 配电网快速高容错性故障定位的线性整数规划方法;郭壮志;徐其兴;洪俊杰;毛晓明;;中国电机工程学报(第03期);全文 * |
| 配电网故障区段定位的互补约束新模型与算法;郭壮志;徐其兴;洪俊杰;孟安波;毛晓明;陈华;;中国电机工程学报(第14期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106324429A (en) | 2017-01-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106324429B (en) | Power distribution network fault positioning method based on RS-IA data mining | |
| CN109902740B (en) | Re-learning industrial control intrusion detection method based on multi-algorithm fusion parallelism | |
| CN110929843A (en) | Abnormal electricity consumption behavior identification method based on improved deep self-coding network | |
| WO2000007113A1 (en) | Automatic freeway incident detection system using artificial neural networks and genetic algorithms | |
| CN112084877B (en) | NSGA-NET-based remote sensing image recognition method | |
| CN112217674B (en) | Alarm root cause identification method based on causal network mining and graph attention network | |
| CN104200096A (en) | Lightning arrester grading ring optimization method based on differential evolutionary algorithm and BP neural network | |
| CN110807518A (en) | Outlier detection method for power grid data | |
| CN113485863B (en) | Method for generating heterogeneous imbalance fault samples based on improved generation of countermeasure network | |
| CN105550578A (en) | Network anomaly classification rule extracting method based on feature selection and decision tree | |
| CN114528949A (en) | Parameter optimization-based electric energy metering abnormal data identification and compensation method | |
| CN110442143A (en) | A kind of unmanned plane situation data clustering method based on combination multiple target dove group's optimization | |
| CN107102223A (en) | NPC photovoltaic DC-to-AC converter method for diagnosing faults based on improved hidden Markov model GHMM | |
| CN108108883B (en) | Clustering algorithm-based vehicle scheduling network elastic simplification method | |
| CN115345222A (en) | Fault classification method based on TimeGAN model | |
| CN106919504B (en) | Test data evolution generation method based on GA algorithm | |
| CN108594079A (en) | A kind of electrical power distribution network fault location method, device and computer-readable medium | |
| CN114842371A (en) | Unsupervised video anomaly detection method | |
| CN113536508A (en) | Method and system for classifying manufacturing network nodes | |
| CN111193254B (en) | Residential daily electricity load prediction method and device | |
| CN117477544A (en) | LSTM ultra-short-term photovoltaic power prediction method and system integrating time mode features | |
| CN104156462A (en) | Complex network community mining method based on cellular automatic learning machine | |
| CN115409122A (en) | Method, system, equipment and medium for analyzing concurrent faults of power transformation equipment | |
| Xu et al. | A Correlation Analysis Model of Fault Location of Distribution System Based on RS-IA Data Mining | |
| CN113986673A (en) | Fault correlation method and device and computer readable storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |