CN105930861A - Adaboost algorithm based transformer fault diagnosis method - Google Patents
Adaboost algorithm based transformer fault diagnosis method Download PDFInfo
- Publication number
- CN105930861A CN105930861A CN201610227946.0A CN201610227946A CN105930861A CN 105930861 A CN105930861 A CN 105930861A CN 201610227946 A CN201610227946 A CN 201610227946A CN 105930861 A CN105930861 A CN 105930861A
- Authority
- CN
- China
- Prior art keywords
- transformer
- weight
- training sample
- basis function
- adaboost algorithm
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000003745 diagnosis Methods 0.000 title claims abstract description 54
- 238000012549 training Methods 0.000 claims abstract description 60
- 238000012360 testing method Methods 0.000 claims abstract description 13
- 238000013528 artificial neural network Methods 0.000 claims description 26
- 239000007789 gas Substances 0.000 claims description 20
- 230000001537 neural effect Effects 0.000 claims description 20
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 17
- 230000000694 effects Effects 0.000 claims description 13
- HSFWRNGVRCDJHI-UHFFFAOYSA-N alpha-acetylene Natural products C#C HSFWRNGVRCDJHI-UHFFFAOYSA-N 0.000 claims description 10
- OTMSDBZUPAUEDD-UHFFFAOYSA-N Ethane Chemical compound CC OTMSDBZUPAUEDD-UHFFFAOYSA-N 0.000 claims description 8
- VGGSQFUCUMXWEO-UHFFFAOYSA-N Ethene Chemical compound C=C VGGSQFUCUMXWEO-UHFFFAOYSA-N 0.000 claims description 8
- 125000002534 ethynyl group Chemical group [H]C#C* 0.000 claims description 8
- 229910052739 hydrogen Inorganic materials 0.000 claims description 8
- 239000001257 hydrogen Substances 0.000 claims description 8
- 238000003064 k means clustering Methods 0.000 claims description 5
- 206010020843 Hyperthermia Diseases 0.000 claims description 4
- 238000010891 electric arc Methods 0.000 claims description 4
- 230000036031 hyperthermia Effects 0.000 claims description 4
- 230000004913 activation Effects 0.000 claims description 3
- 230000013011 mating Effects 0.000 claims description 3
- 239000012141 concentrate Substances 0.000 claims description 2
- 125000004435 hydrogen atom Chemical class [H]* 0.000 claims 1
- 230000006870 function Effects 0.000 description 68
- 238000012544 monitoring process Methods 0.000 description 10
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 description 8
- 230000008859 change Effects 0.000 description 5
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 description 4
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 description 4
- 229910002092 carbon dioxide Inorganic materials 0.000 description 4
- 239000001569 carbon dioxide Substances 0.000 description 4
- 229910002091 carbon monoxide Inorganic materials 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000012423 maintenance Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 3
- 150000002431 hydrogen Chemical class 0.000 description 3
- 230000006978 adaptation Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 241001269238 Data Species 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000000470 constituent Substances 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000002405 diagnostic procedure Methods 0.000 description 1
- 238000009413 insulation Methods 0.000 description 1
- 230000007257 malfunction Effects 0.000 description 1
- 230000007935 neutral effect Effects 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000004613 tight binding model Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/245—Classification techniques relating to the decision surface
- G06F18/2453—Classification techniques relating to the decision surface non-linear, e.g. polynomial classifier
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Probability & Statistics with Applications (AREA)
- Mathematical Physics (AREA)
- Nonlinear Science (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to an Adaboost algorithm based transformer fault diagnosis method. The method comprises that a training sample set is used to train weak classifiers; the weak classifiers are integrated into a strong classifier of higher classification precision after cycle training and weight adjustment; and a test sample serves as input of the strong classifier to obtain the corresponding fault type. According to the invention, the weak classifiers are integrated to solve the problem that the strong classifier is hard to obtain, operation is simple, the transformer fault mode is identified in the classified manner, and the practicality is high.
Description
Technical field
The invention belongs to transformer fault on-line monitoring technique field, refer in particular to the gas according to producing during transformer fault and transformer is carried out fault diagnosis, be specifically related to a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm.
Background technology
Along with the high speed development of power grid construction, China's electrical network, from city island network, develops into large regional grid, transferring electricity from the west to the east, north and south confession mutually, and on national network general layout is formed.Power system be one by numerous, give, defeated, join, big system that electrical equipment is formed by connecting, the reliability of these equipment and operation conditions directly determine the stable of whole system and safety, also determine power supply quality and power supply reliability, along with power system is to high voltage, Large Copacity, internet development, and the raising of each electricity consumption goal, the requirement to the security reliability index of power system is more and more higher.Power transformer is the part that power system is important, and its normal reliable runs the one of the main reasons being to ensure that whole power grid operation.The preventative maintenance system formed in decades, has played very important effect to the reliability improving operation of power networks, but this can not insulation hidden danger within discovering device in time.
It addition, the expense of preventative maintenance is the highest.Along with electrical network develops more and more higher to the requirement of power supply reliability with national economy to increasingly automated direction, in the urgent need to changing existing maintenance of equipment system, the CBM System Based based on on-line monitoring and fault diagnosis technology gradually replaces preventative maintenance system or the development trend for tracing and monitoring fault is the clearest and the most definite.The most reliably the diagnosing malfunction tool that transformer is potential is of great significance.
Method of Fault Diagnosis in Transformer has a variety of, such as at present: BP neutral net is that the troubleshooting issue of transformer provides a kind of reasonable structural system, but there is the shortcoming that convergence speed is slow, be easily trapped into local minimum point；Expert system simulated failure diagnosing human expert can complete failure diagnostic process effectively, but there is also many technical problems such as knowledge acquisition difficulty, uncertain inference and self study difficulty；Fuzzy control can by accurate mathematical tool by fuzzy concept or natural language sharpening, thus to phenomenon of the failure can quantization etc. the most in addition, but its fuzzy membership function needs expertise or repetition test just to can determine that.
In order to solve the problems referred to above, the present invention proposes a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm.
Summary of the invention
For the deficiency of the problems referred to above, the present invention proposes a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm.It is an object of the invention to solve at least one the problems referred to above or defect, and at least one advantage that will be described later is provided.
It is a still further object of the present invention to provide a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm, by utilizing AdaBoost that described radial basis function neural network grader is constantly trained, constantly adjust according to error, it is combined by Nearest Neighbor with Weighted Voting again and is promoted to final strong classifier, improve fault diagnosis precision.
It is a still further object of the present invention to provide a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm, the gas on-site data that its transformer online monitoring system is monitored in real time are as sample set data, change in concentration according to these transformer fault characteristic gas, it was predicted that go out transformer fault mode type.
It is a still further object of the present invention to provide a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm, AdaBoost algorithm is applied on transformer fault diagnosis is analyzed by it, predicts transformer fault according to the change in concentration of gas on-site.
It is a still further object of the present invention to provide a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm, it greatly reduces computation complexity and amount of calculation, be more suitable for online quick diagnosis.
In order to realize according to object of the present invention and further advantage, the invention provides a kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm, comprise the following steps:
Step one, gather the concentration value of multiple transformer fault characteristic gas, as the sample of sample set after multiple described concentration value is normalized, encode the fault mode type of transformer simultaneously, using the target output coding of the fault mode type of corresponding transformer as class label corresponding to property value in described sample set, using a part of sample set as training sample set, and another part sample set is as test sample collection；
Wherein, each training sample that described training sample is concentrated has equal initial weight；
Step 2, uses radial basis function neural network to be set to the center of described radial basis function neural network as Weak Classifier, all samples concentrated by described training sample；
Step 3, is circulated training, until obtaining anticipation function by radial basis function neural network described in described training sample set pair；
Step 4, according to the prediction effect of described anticipation function, gives the weight that described anticipation function is different from described initial weight, uses the ballot method of Weight to produce final anticipation function combined sequence；
Step 5, inputs the described test sample collection in described step one in described final anticipation function combined sequence, after described final anticipation function combined sequence judges to identify, by mating described class label, draws described fault mode type, completes diagnosis.
Adaboost algorithm only need to find the Weak Classifier of several classification accuracies more slightly higher than random assortment accuracy rate (i.e. accuracy is slightly larger than 50%), the present invention selects described radial basis function neural network, by its training is produced anticipation function sequence, and use Nearest Neighbor with Weighted Voting mechanism to produce described final anticipation function, it is final strong classifier.It is thus possible to Weak Classifier to be organically integrated into a higher described final strong classifier of nicety of grading.Utilize the method that transformer fault mode type is carried out Classification and Identification and there is preferable practicality.The present invention is by doing contrast experiment's discovery, the classification accuracy using single radial base neural net algorithm is about 81.25%, and using the classification accuracy of Diagnosis Method of Transformer Faults method based on Adaboost algorithm is 93.75%, overall accuracy rate improves 12.5%.
Preferably, in described step one, the fault type simultaneously encoding transformer is specific as follows:
Overheated to normal condition, middle temperature, hyperthermia and superheating, shelf depreciation, spark discharge and arc discharge six class fault mode are separately encoded into Arabic numerals 1,2,3,4,5 and 6, as the class label corresponding to property value in described sample set.
Adaboost algorithm is applied in the actual application of transformer fault mode type identification by the present invention, improves transformer fault accuracy of identification.
Preferably, described transformer fault characteristic gas is hydrogen, methane, ethane, ethene and acetylene.
The data of the present invention derive from the gas on-site data that transformer online monitoring system is monitored in real time, the information such as hydrogen that not only monitoring transformer produces when breaking down, methane, ethane, ethene, acetylene, carbon monoxide, carbon dioxide, and substantially dope transformer fault according to the change in concentration of these gases.
Preferably, in described step one, by initializing the weight coefficient of each training sample that described training sample is concentrated so that each described sample has equal initial weight.
Preferably, in described step 2, use radial basis function neural network as Weak Classifier, need to be activated by Green's activation primitive.
Preferably, in described step 3, it is circulated training by radial basis function neural network described in described training sample set pair, until obtaining specifically comprising the following steps that of anticipation function
The final Basis Function Center of described radial basis function neural network is sought based on K-means clustering method；
According to the ultimate range between described training sample and described final Basis Function Center, obtain variance yields；
By described final Basis Function Center and described variance yields, obtain hidden layer in described radial base neural net to the connection weights between output layer；
Described connection weights are combined the structure of described radial base neural net, obtain the actual output of described radial base neural net, the positive negative difference that described actual output and target export is in preset range, then training terminates, obtain anticipation function, described positive negative difference not in described preset range, then repeats above step；
Wherein, the described actual actual output coding being output as corresponding transformer fault mode type, described target is output as the target output coding of corresponding transformer fault mode type.
Preferably, the final Basis Function Center of described radial basis function neural network is asked to comprise the following steps based on K-means clustering method:
A, choose n training sample that described training sample concentrates as cluster centre, according to the Euclidean distance between input sample and described cluster centre, described input sample is assigned to during each cluster gathers；Calculate the mean value of training sample in each described cluster set, obtain new cluster centre, it is judged that the described new cluster centre center deciding whether to carry out next round that whether changes solves；It is then to perform step B, no, then perform step C；
B, the described new cluster centre of above-mentioned steps are the final Basis Function Center of radial basis function neural network；
C, again choose n described training sample as cluster centre, enter the solving of described cluster centre of next round；
Wherein, input sample is the sample that described training sample is concentrated, and described input sample is the classification that the Euclidean distance between basis and center is carried out.
Preferably, in described step 4, give its weight different from described initial weight according to the prediction effect of described anticipation function sequence, use the ballot method of Weight to produce specifically comprising the following steps that of final anticipation function combined sequence
Calculate the weight training error of described anticipation function, obtain the prediction effect of described anticipation function；
According to described prediction effect, give described anticipation function sequence the first weight, and described first weight is updated, obtain the second weight；
Second weight is normalized, and sums up normalized respective weights, produces final anticipation function combined sequence by ballot method.
When the prediction effect (training error) of described anticipation function is more than 0.5, the second weight of described final anticipation function is normalized, and normalized respective weights is summed up, produce final anticipation function combined sequence by ballot method.
Adaboost algorithm only need to find the Weak Classifier of several classification accuracies more slightly higher than random assortment accuracy rate (i.e. accuracy is slightly larger than 50%), anticipation function sequence is produced by training, and use Nearest Neighbor with Weighted Voting mechanism to produce final anticipation function, such that it is able to Weak Classifier to be organically integrated into a higher strong classifier of nicety of grading, the most described final anticipation function, utilize the method that transformer fault pattern is carried out Classification and Identification and there is preferable practicality, solve the problem that strong classifier is difficult to obtain, also can improve accuracy of identification simultaneously.
Preferably, described preset range is 0.01-0.05.
Preset range is typically set between 0-1, but the described preset range of the present invention is narrower so that the error of identification transformer fault mode type of the present invention is little, and the degree of accuracy is higher.
Beneficial effects of the present invention:
1, the Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, is trained radial basis function neural network grader, then is combined by Nearest Neighbor with Weighted Voting and is promoted to final strong classifier, improves fault diagnosis precision.
2, the Diagnosis Method of Transformer Faults of based on Adaboost algorithm that the present invention provides, during Fault Diagnosis Method of Power Transformer mode type, it is to avoid the impact of the subjective factor of people, makes selection more objective, and classification accuracy rate is higher.Experiment shows, the classification accuracy of Fault Diagnosis Method of Power Transformer types of models of the present invention is 93.75%, improves 12.5% than the classification accuracy of single radial base neural net algorithm.
3, the Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, by integrated weak learning algorithm, solves the problem that strong learning algorithm is difficult to obtain, also can improve precision of prediction simultaneously.
4, the Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, use radial base neural net (RBF) as Weak Classifier, RBF has that accuracy rate is high, Adaptation of structure should determine that, export and good characteristic that initial weight is unrelated, the time making training network is far smaller than other network training, shortens the time of Fault Diagnosis Method of Power Transformer types of models.
5, the Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, the sample data of this analysis method derives from the transformer fault characteristic gas that transformer online monitoring system is monitored in real time, the hydrogen of generation, methane, ethane, ethene, acetylene, carbon monoxide, carbon dioxide isoconcentration information when not only monitoring transformer breaks down, and substantially dope transformer fault according to the change in concentration of these gases.
6, the Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, it can be greatly reduced calculating cost, meet the requirement of online Fast Classification diagnosis.
Accompanying drawing explanation
Fig. 1 is radial base neural net structural representation of the present invention；
Fig. 2 is the structure chart of described Diagnosis Method of Transformer Faults based on Adaboost algorithm；
Fig. 3 is the flow chart of described Diagnosis Method of Transformer Faults based on Adaboost algorithm.
Detailed description of the invention
The present invention will be further described with detailed description of the invention below in conjunction with the accompanying drawings.
The flow chart of the present invention as shown in Figure 3, comprises the following steps:
Step one, gather the concentration value of multiple transformer fault characteristic gas, as the sample of sample set after multiple described concentration value is normalized, encode the fault mode type of transformer simultaneously, overheated to normal condition, middle temperature, hyperthermia and superheating, shelf depreciation, spark discharge and arc discharge six class fault mode are separately encoded as Arabic numerals 1,2,3,4,5 and 6, as the class label corresponding to property value in described sample set, using a part of sample set as training sample set, and another part sample set is as test sample collection；
Wherein, described transformer fault characteristic gas is hydrogen, methane, ethane, ethene and acetylene, it is also possible to for the combination of hydrogen, methane, ethane, ethene, acetylene, carbon monoxide and carbon dioxide；
Training sample S={ (x1, y1) after being normalized, (x2, y2) ..., (xn, yn) }, wherein, xi ∈ X, (concentration of 5 kinds of characteristic gas when xi is transformer fault), yi ∈ Y={1,2 ..., k}, the fault type of corresponding transformer.Initialize the weight coefficient of each sample simultaneouslyI=1,2 ..., n.Each sample i.e. has equal initial weight.
The sample data of this analysis method derives from the transformer fault characteristic gas that transformer online monitoring system is monitored in real time, the hydrogen of generation, methane, ethane, ethene, acetylene, carbon monoxide, carbon dioxide isoconcentration information when not only monitoring transformer breaks down, and substantially dope transformer fault according to the change in concentration of these gases.
Step 2, uses radial basis function neural network to be set to the center of described radial basis function neural network as Weak Classifier, all samples concentrated by described training sample；
Use radial basis function neural network RBFNN as Weak Classifier, first construct RBF neural.Choosing Hidden nodes equal to inputting sample number, the activation primitive of hidden node is Green function, and concrete functional expression is:
In formula, | | xn-ci| | for European norm, c is the center of Green's function, and σ is the variance of Green's function.All input samples are set to the center of RBF simultaneously, and each RBF takes unified extension constant.Network structure is as shown in Figure 1.
Use radial base neural net (RBF) as Weak Classifier, RBF has that accuracy rate is high, Adaptation of structure should determine that, export and good characteristic that initial weight is unrelated, the time making training network is far smaller than other network training, shortens the time of Fault Diagnosis Method of Power Transformer types of models.
Step 3, is circulated training, until it is specific as follows to obtain anticipation function by radial basis function neural network described in described training sample set pair:
The final Basis Function Center of described radial basis function neural network is sought based on K-means clustering method；Choose n given training sample as cluster centre ci (i=1,2 ..., n)；According to xp and center be the Euclidean distance between ci xp is assigned to input sample each cluster set θ p (p=1,2 ..., P) in；Calculate the mean value of training sample in each cluster set θ p, the newest cluster centre ci, if new cluster centre no longer changes, then the ci of gained is the Basis Function Center that RBF neural is final, otherwise redistributing training sample set, the center entering next round solves.
According to the ultimate range between described training sample and described final Basis Function Center, obtaining variance yields, solve variances sigma i, formula is as follows
In formula, cmax is the ultimate range between selected center, i=1,2 ..., n
By described final Basis Function Center and described variance yields, obtaining hidden layer in described radial base neural net to the connection weights between output layer, the formula calculating weights between hidden layer and output layer is as follows:
Wherein, n=1,2 ..., n；I=1,2 ..., n
Described connection weights are combined the structure of described radial base neural net, obtains the actual output y of described radial base neural netj, the positive negative difference that described actual output and target export is in preset range, then training terminates, and obtains anticipation function, and described positive negative difference not in described preset range, then repeats above step；The formula of described actual output is as follows:
Wherein, j=1,2 ..., in n formula,It is the n-th input sample, n=1,2 ..., N, N represent total sample number, and ci is the center of network hidden layer node, and ω ij is the hidden layer connection weights to output layer, i=1,2 ..., n is the nodal point number of hidden layer.
Wherein, the described actual actual output coding being output as corresponding transformer fault mode type, described target is output as the target output coding of corresponding transformer fault mode type.
Wherein, described preset range given to this invention is 0.01-0.05.
The Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, it can be greatly reduced calculating cost, meet the requirement of online Fast Classification diagnosis.
Step 4, according to the prediction effect of described anticipation function, gives the weight that described anticipation function is different from described initial weight, uses the ballot method of Weight to produce final anticipation function combined sequence specific as follows:
Calculate the weight training error of described anticipation function, obtain the prediction effect of described anticipation function；Calculating the weight training error of ht, be exactly i.e. wrong point rate, formula is as follows:
Wherein, if yi≠ht(xi), then I=1；Otherwise, I=0.
According to described prediction effect, give described anticipation function sequence the first weight α(t), and described first weight is updated, obtain the second weight；
Wherein, the formula of described first weight is as follows:
And the renewal of the first weight is carried out by above formula, the formula updating coefficient is:
Wherein, ZtFor normalization coefficient, can make
When the training error of anticipation function is more than 0.5, described the second weight finally giving anticipation function is normalized, obtains strong classifier:
Work as htDuring (x)=k, corresponding weights are summed up, by the ballot method final anticipation function combined sequence of generation:
Wherein, arg max g (t), expression is the subset defining territory, and in this subset, either element all can make function g () take maximum.
The Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, radial basis function neural network grader is trained, it is combined by Nearest Neighbor with Weighted Voting again and is promoted to final strong classifier, improve fault diagnosis precision, that is by integrated weak learning algorithm, solve the problem that strong learning algorithm is difficult to obtain, also can improve precision of prediction simultaneously.
Step 5, inputs the described test sample collection in described step one in described final anticipation function combined sequence, after described final anticipation function combined sequence judges to identify, by mating described class label, draws described fault mode type, completes diagnosis.
The Diagnosis Method of Transformer Faults based on Adaboost algorithm that the present invention provides, is trained radial basis function neural network grader, then is combined by Nearest Neighbor with Weighted Voting and is promoted to final strong classifier, improves fault diagnosis precision.
On the other hand, the present invention can be greatly reduced calculating cost, meets the requirement of online Fast Classification diagnosis.
The embodiment that the present invention provides, choose hydrogen (H2), methane (CH4), ethane (C2H6), ethene (C2H4), acetylene (C2H2) these five kinds of characteristic feature gases component as primitive attribute data, first dissolved gas constituent content is normalized, make it all [-1,1] in the range of, as the property value in sample set.
Then encoding transformer fault type, overheated to normal condition, middle temperature, hyperthermia and superheating, shelf depreciation, spark discharge, arc discharge is separately encoded is 1,2,3,4,5,6, as the class label corresponding to property value in sample set.
Use the 305 groups of Gases Dissolved in Transformer Oil components having determined that fault type as training and test sample.Concentrating at training sample, each fault type has 35 groups of sample datas, remaining 95 groups of test sample as model.The Diagnosis Method of Transformer Faults utilizing 210 groups of data that this patent proposes Adaboost algorithm analyzes being trained of model, uses 95 groups of data to test, and wherein, part described test sample collection is 16 groups, as shown in table 1.
16 groups of data of test sample collection described in table 1 part
In order to verify effectiveness of the invention and accuracy, and performance is good and bad compared with the classification of single radial base neural net algorithm, has carried out one group of contrast experiment, the accuracy of two kinds of sorting techniques such as table 2.
Table 2 present invention and the comparative result of single radial base neural net algorithm classification accuracy
Can be drawn by table 2, experiment finds that the classification accuracy using single radial base neural net algorithm is about 81.25%, and using the Diagnosis Method of Transformer Faults of Adaboost algorithm to analyze category of model accuracy rate is 93.75%, the method of the invention is compared with single radial base neural net algorithm, and classification accuracy improves 12.5%.
The present invention uses the Adaboost algorithm of transformer fault figure penalties function based on Adaboost algorithm.Adaboost is that a kind of realization simply, applies the simplest algorithm, and will not over-fitting.Additionally, this algorithm of Adaboost algorithm based on figure penalties function is substantially reduced the required precision to Weak Classifier, and algorithm is simple and clear, can direct solution multicategory classification problem, it is greatly reduced computation complexity and amount of calculation, is a kind of algorithm being suitably applied transformer fault diagnosis.
The present invention also has the data of other embodiments at this, does not enumerates.
Although embodiment of the present invention are disclosed as above, but it is not restricted in specification and embodiment listed utilization, it can be applied to various applicable the field of the invention completely, for those skilled in the art, it is easily achieved other amendment, therefore, under the universal limited without departing substantially from claim and equivalency range, the present invention is not limited to specific details and shown here as the legend with description.
Claims (9)
1. a Diagnosis Method of Transformer Faults based on Adaboost algorithm, it is characterised in that comprise the following steps:
Step one, gather the concentration value of multiple transformer fault characteristic gas, as the sample of sample set after multiple described concentration value is normalized, encode the fault mode type of transformer simultaneously, using the target output coding of the fault mode type of corresponding transformer as class label corresponding to property value in described sample set, using a part of sample set as training sample set, and another part sample set is as test sample collection；
Wherein, each training sample that described training sample is concentrated has equal initial weight；
Step 2, uses radial basis function neural network to be set to the center of described radial basis function neural network as Weak Classifier, all samples concentrated by described training sample；
Step 3, is circulated training, until obtaining anticipation function by radial basis function neural network described in described training sample set pair；
Step 4, according to the prediction effect of described anticipation function, gives the weight that described anticipation function is different from described initial weight, uses the ballot method of Weight to produce final anticipation function combined sequence；
Step 5, inputs the described test sample collection in described step one in described final anticipation function combined sequence, after described final anticipation function combined sequence judges to identify, by mating described class label, draws described fault mode type, completes diagnosis.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 1, it is characterised in that in described step one, the fault type simultaneously encoding transformer is specific as follows:
Overheated to normal condition, middle temperature, hyperthermia and superheating, shelf depreciation, spark discharge and arc discharge six class fault mode are separately encoded into Arabic numerals 1,2,3,4,5 and 6, as the class label corresponding to property value in described sample set.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 1, it is characterised in that multiple described transformer fault characteristic gas is hydrogen, methane, ethane, ethene and acetylene.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 1, it is characterized in that, in described step one, by initializing the weight coefficient of each training sample that described training sample is concentrated so that each described training sample has equal initial weight.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most as claimed in any of claims 2 to 4, it is characterised in that in described step 2, is used radial basis function neural network as Weak Classifier, need to be activated by Green's activation primitive.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 1, it is characterized in that, in described step 3, it is circulated training by radial basis function neural network described in described training sample set pair, until obtaining specifically comprising the following steps that of anticipation function
The final Basis Function Center of described radial basis function neural network is sought based on K-means clustering method；
According to the ultimate range between described training sample and described final Basis Function Center, obtain variance yields；
By described final Basis Function Center and described variance yields, obtain hidden layer in described radial base neural net to the connection weights between output layer；
Described connection weights are combined the structure of described radial base neural net, obtain the actual output of described radial base neural net, the positive negative difference that described actual output and target export is in preset range, then training terminates, obtain anticipation function, described positive negative difference not in described preset range, then repeats above step；
Wherein, the described actual actual output coding being output as corresponding transformer fault mode type, described target is output as the target output coding of corresponding transformer fault mode type.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 6, it is characterised in that ask the final Basis Function Center of described radial basis function neural network to comprise the following steps based on K-means clustering method:
A, choose n training sample that described training sample concentrates as cluster centre, according to the Euclidean distance between input sample and described cluster centre, described input sample is assigned to during each cluster gathers；Calculate the mean value of training sample in each described cluster set, obtain new cluster centre, it is judged that the described new cluster centre center deciding whether to carry out next round that whether changes solves；It is then to perform step B, no, then perform step C；
B, the described new cluster centre of above-mentioned steps are the final Basis Function Center of described radial basis function neural network；
C, again choose n described training sample as cluster centre, enter the solving of described cluster centre of next round.
Diagnosis Method of Transformer Faults based on Adaboost algorithm the most according to claim 1, it is characterized in that, in described step 4, prediction effect according to described anticipation function gives the weight that they are different from described initial weight, uses the ballot method of Weight to produce specifically comprising the following steps that of final anticipation function combined sequence
Calculate the weight training error of described anticipation function, obtain the prediction effect of described anticipation function；
According to described prediction effect, give described anticipation function the first weight, and described first weight is updated, obtain the second weight；
Described second weight is normalized, and normalized respective weights is summed up, produce final anticipation function combined sequence by ballot method.
9. according to the Diagnosis Method of Transformer Faults based on Adaboost algorithm described in claim 6, it is characterised in that described preset range is 0.01-0.05.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610227946.0A CN105930861B (en) | 2016-04-13 | 2016-04-13 | A kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610227946.0A CN105930861B (en) | 2016-04-13 | 2016-04-13 | A kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105930861A true CN105930861A (en) | 2016-09-07 |
| CN105930861B CN105930861B (en) | 2019-08-06 |
Family
ID=56837998
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610227946.0A Active CN105930861B (en) | 2016-04-13 | 2016-04-13 | A kind of Diagnosis Method of Transformer Faults based on Adaboost algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105930861B (en) |
Cited By (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106501693A (en) * | 2016-12-08 | 2017-03-15 | 贵州电网有限责任公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults based on fuzzy Boltzmann machine |
| CN106530082A (en) * | 2016-10-25 | 2017-03-22 | 清华大学 | Stock predication method and stock predication system based on multi-machine learning |
| CN107180140A (en) * | 2017-06-08 | 2017-09-19 | 中南大学 | Shafting fault recognition method based on dual-tree complex wavelet and AdaBoost |
| CN107194465A (en) * | 2017-06-16 | 2017-09-22 | 华北电力大学（保定） | A kind of method that utilization virtual sample trains Neural Network Diagnosis transformer fault |
| CN108021945A (en) * | 2017-12-07 | 2018-05-11 | 广东电网有限责任公司电力科学研究院 | A kind of transformer state evaluation model method for building up and device |
| CN108717149A (en) * | 2018-05-25 | 2018-10-30 | 西安工程大学 | Diagnosis Method of Transformer Faults based on M-RVM fusion dynamic weightings AdaBoost |
| CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
| CN109188162A (en) * | 2018-07-17 | 2019-01-11 | 广东工业大学 | It is a kind of based on the Transformer condition evaluation that can open up radial base neural net |
| CN109325519A (en) * | 2018-08-20 | 2019-02-12 | 中国铁道科学研究院集团有限公司电子计算技术研究所 | Fault recognition method and device |
| CN109726767A (en) * | 2019-01-13 | 2019-05-07 | 胡燕祝 | A kind of perceptron network data classification method based on AdaBoost algorithm |
| CN109858564A (en) * | 2019-02-21 | 2019-06-07 | 上海电力学院 | Modified Adaboost-SVM model generating method suitable for wind electric converter fault diagnosis |
| CN110163531A (en) * | 2019-06-02 | 2019-08-23 | 南京邮电大学盐城大数据研究院有限公司 | Network transformer abnormality method for early warning based on K- cluster |
| CN110263837A (en) * | 2019-06-13 | 2019-09-20 | 河海大学 | A kind of circuit breaker failure diagnostic method based on multilayer DBN model |
| CN110286161A (en) * | 2019-03-28 | 2019-09-27 | 清华大学 | Main transformer method for diagnosing faults based on adaptive enhancing study |
| CN110289097A (en) * | 2019-07-02 | 2019-09-27 | 重庆大学 | A kind of Pattern Recognition Diagnosis system stacking model based on Xgboost neural network |
| CN110618340A (en) * | 2019-03-11 | 2019-12-27 | 广东工业大学 | Transformer state evaluation method |
| CN111723518A (en) * | 2020-05-29 | 2020-09-29 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis device and method based on condition inference tree and AdaBoost |
| CN111767675A (en) * | 2020-06-24 | 2020-10-13 | 国家电网有限公司大数据中心 | Transformer vibration fault monitoring method and device, electronic equipment and storage medium |
| CN111860658A (en) * | 2020-07-24 | 2020-10-30 | 华北电力大学（保定） | Transformer fault diagnosis method based on cost sensitivity and integrated learning |
| US20210056246A1 (en) * | 2019-08-21 | 2021-02-25 | Northwestern Polytechnical University | Method for evaluating reliability of a sealing structure in a multi-failure mode based on an adaboost algorithm |
| CN112580715A (en) * | 2020-12-16 | 2021-03-30 | 珠海格力电器股份有限公司 | Household equipment fault detection method, device, equipment and medium |
| CN112710956A (en) * | 2020-12-17 | 2021-04-27 | 四川虹微技术有限公司 | Battery management system fault detection system and method based on expert system |
| CN113391172A (en) * | 2021-05-31 | 2021-09-14 | 国网山东省电力公司电力科学研究院 | Partial discharge diagnosis method and system based on time sequence integration and used for multi-source ultrasonic detection |
| CN113705405A (en) * | 2021-08-19 | 2021-11-26 | 电子科技大学 | Nuclear pipeline fault diagnosis method |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5640103A (en) * | 1994-06-30 | 1997-06-17 | Siemens Corporate Research, Inc. | Radial basis function neural network autoassociator and method for induction motor monitoring |
| CN103207950A (en) * | 2013-04-16 | 2013-07-17 | 郑州航空工业管理学院 | Intelligent transformer fault diagnostic method based on RBF (radial basis function) neural network |
| CN103235973A (en) * | 2013-04-16 | 2013-08-07 | 郑州航空工业管理学院 | Transformer fault diagnosis method based on radial basis function neural network |
| CN104299035A (en) * | 2014-09-29 | 2015-01-21 | 国家电网公司 | Method for diagnosing fault of transformer on basis of clustering algorithm and neural network |
| CN104616033A (en) * | 2015-02-13 | 2015-05-13 | 重庆大学 | Fault diagnosis method for rolling bearing based on deep learning and SVM (Support Vector Machine) |
-
2016
- 2016-04-13 CN CN201610227946.0A patent/CN105930861B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5640103A (en) * | 1994-06-30 | 1997-06-17 | Siemens Corporate Research, Inc. | Radial basis function neural network autoassociator and method for induction motor monitoring |
| CN103207950A (en) * | 2013-04-16 | 2013-07-17 | 郑州航空工业管理学院 | Intelligent transformer fault diagnostic method based on RBF (radial basis function) neural network |
| CN103235973A (en) * | 2013-04-16 | 2013-08-07 | 郑州航空工业管理学院 | Transformer fault diagnosis method based on radial basis function neural network |
| CN104299035A (en) * | 2014-09-29 | 2015-01-21 | 国家电网公司 | Method for diagnosing fault of transformer on basis of clustering algorithm and neural network |
| CN104616033A (en) * | 2015-02-13 | 2015-05-13 | 重庆大学 | Fault diagnosis method for rolling bearing based on deep learning and SVM (Support Vector Machine) |
Non-Patent Citations (3)
| Title |
|---|
| B. RAVIKUMAR 等: "Application of support vector machines for fault diagnosis in power transmission system", 《IET GENERATION, TRANSMISSION & DISTRIBUTION》 * |
| 高宏岩: "基于径向基概率神经网络的变压器故障诊断", 《煤矿机械》 * |
| 麻闽政: "径向基函数神经网络在电力变压器故障诊断中的应用", 《广东电力》 * |
Cited By (31)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106530082A (en) * | 2016-10-25 | 2017-03-22 | 清华大学 | Stock predication method and stock predication system based on multi-machine learning |
| CN106501693A (en) * | 2016-12-08 | 2017-03-15 | 贵州电网有限责任公司电力科学研究院 | A kind of Diagnosis Method of Transformer Faults based on fuzzy Boltzmann machine |
| CN107180140A (en) * | 2017-06-08 | 2017-09-19 | 中南大学 | Shafting fault recognition method based on dual-tree complex wavelet and AdaBoost |
| CN107180140B (en) * | 2017-06-08 | 2019-12-10 | 中南大学 | Shafting fault identification method based on dual-tree complex wavelet and AdaBoost |
| CN107194465A (en) * | 2017-06-16 | 2017-09-22 | 华北电力大学（保定） | A kind of method that utilization virtual sample trains Neural Network Diagnosis transformer fault |
| CN108021945A (en) * | 2017-12-07 | 2018-05-11 | 广东电网有限责任公司电力科学研究院 | A kind of transformer state evaluation model method for building up and device |
| CN108717149A (en) * | 2018-05-25 | 2018-10-30 | 西安工程大学 | Diagnosis Method of Transformer Faults based on M-RVM fusion dynamic weightings AdaBoost |
| CN109063734A (en) * | 2018-06-29 | 2018-12-21 | 广东工业大学 | The oil-immersed transformer malfunction appraisal procedure clustered in conjunction with multistage local density |
| CN109063734B (en) * | 2018-06-29 | 2022-02-25 | 广东工业大学 | Oil-immersed transformer fault state evaluation method combining multi-level local density clustering |
| CN109188162A (en) * | 2018-07-17 | 2019-01-11 | 广东工业大学 | It is a kind of based on the Transformer condition evaluation that can open up radial base neural net |
| CN109325519A (en) * | 2018-08-20 | 2019-02-12 | 中国铁道科学研究院集团有限公司电子计算技术研究所 | Fault recognition method and device |
| CN109726767A (en) * | 2019-01-13 | 2019-05-07 | 胡燕祝 | A kind of perceptron network data classification method based on AdaBoost algorithm |
| CN109858564A (en) * | 2019-02-21 | 2019-06-07 | 上海电力学院 | Modified Adaboost-SVM model generating method suitable for wind electric converter fault diagnosis |
| CN109858564B (en) * | 2019-02-21 | 2023-05-05 | 上海电力学院 | Improved Adaboost-SVM model generation method suitable for wind power converter fault diagnosis |
| CN110618340A (en) * | 2019-03-11 | 2019-12-27 | 广东工业大学 | Transformer state evaluation method |
| CN110286161A (en) * | 2019-03-28 | 2019-09-27 | 清华大学 | Main transformer method for diagnosing faults based on adaptive enhancing study |
| CN110163531A (en) * | 2019-06-02 | 2019-08-23 | 南京邮电大学盐城大数据研究院有限公司 | Network transformer abnormality method for early warning based on K- cluster |
| CN110263837A (en) * | 2019-06-13 | 2019-09-20 | 河海大学 | A kind of circuit breaker failure diagnostic method based on multilayer DBN model |
| CN110289097A (en) * | 2019-07-02 | 2019-09-27 | 重庆大学 | A kind of Pattern Recognition Diagnosis system stacking model based on Xgboost neural network |
| US20210056246A1 (en) * | 2019-08-21 | 2021-02-25 | Northwestern Polytechnical University | Method for evaluating reliability of a sealing structure in a multi-failure mode based on an adaboost algorithm |
| US11657335B2 (en) * | 2019-08-21 | 2023-05-23 | Northwestern Polytechnical University | Method for evaluating reliability of a sealing structure in a multi-failure mode based on an adaboost algorithm |
| CN111723518A (en) * | 2020-05-29 | 2020-09-29 | 国网四川省电力公司电力科学研究院 | Transformer fault diagnosis device and method based on condition inference tree and AdaBoost |
| CN111767675A (en) * | 2020-06-24 | 2020-10-13 | 国家电网有限公司大数据中心 | Transformer vibration fault monitoring method and device, electronic equipment and storage medium |
| CN111860658A (en) * | 2020-07-24 | 2020-10-30 | 华北电力大学（保定） | Transformer fault diagnosis method based on cost sensitivity and integrated learning |
| CN112580715A (en) * | 2020-12-16 | 2021-03-30 | 珠海格力电器股份有限公司 | Household equipment fault detection method, device, equipment and medium |
| CN112580715B (en) * | 2020-12-16 | 2024-05-07 | 珠海格力电器股份有限公司 | Household equipment fault detection method, device, equipment and medium |
| CN112710956A (en) * | 2020-12-17 | 2021-04-27 | 四川虹微技术有限公司 | Battery management system fault detection system and method based on expert system |
| CN112710956B (en) * | 2020-12-17 | 2023-08-04 | 四川虹微技术有限公司 | Expert system-based battery management system fault detection system and method |
| CN113391172A (en) * | 2021-05-31 | 2021-09-14 | 国网山东省电力公司电力科学研究院 | Partial discharge diagnosis method and system based on time sequence integration and used for multi-source ultrasonic detection |
| CN113705405A (en) * | 2021-08-19 | 2021-11-26 | 电子科技大学 | Nuclear pipeline fault diagnosis method |
| CN113705405B (en) * | 2021-08-19 | 2023-04-18 | 电子科技大学 | Nuclear pipeline fault diagnosis method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105930861B (en) | 2019-08-06 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105930861A (en) | Adaboost algorithm based transformer fault diagnosis method | |
| CN105930901B (en) | A kind of Diagnosis Method of Transformer Faults based on RBPNN | |
| Thai-Nghe et al. | Factorization Models for Forecasting Student Performance. | |
| CN101063643B (en) | Intelligent diagnostic method for airplane functional failure and system thereof | |
| CN104155574A (en) | Power distribution network fault classification method based on adaptive neuro-fuzzy inference system | |
| Feng et al. | Analysis and prediction of students’ academic performance based on educational data mining | |
| CN109931678A (en) | Air-conditioning fault diagnosis method based on deep learning LSTM | |
| Akram et al. | Improving stealth assessment in game-based learning with LSTM-based analytics | |
| CN109242149A (en) | A kind of student performance early warning method and system excavated based on educational data | |
| CN102289682B (en) | Transformer fault diagnosis method based on integrated learning Bagging algorithm | |
| CN101404071B (en) | Electronic circuit fault diagnosis neural network method based on grouping particle swarm algorithm | |
| CN105574589B (en) | Transformer oil chromatographic method for diagnosing faults based on niche genetic algorithm | |
| CN103995237A (en) | Satellite power supply system online fault diagnosis method | |
| CN101587155A (en) | Oil soaked transformer fault diagnosis method | |
| CN107085763A (en) | A kind of driving motor for electric automobile system performance evaluation method | |
| CN109858797A (en) | The various dimensions information analysis of the students method of knowledge based network exact on-line education system | |
| CN110542819A (en) | transformer fault type diagnosis method based on semi-supervised DBNC | |
| CN106709192A (en) | Power distribution network three-dimensional simulation training credibility evaluation method based on cloud matter-element model | |
| CN107545307A (en) | Predicting model for dissolved gas in transformer oil method and system based on depth belief network | |
| CN106597154B (en) | Transformer fault diagnosis method for improving based on DAG-SVM | |
| Wang et al. | Design and implementation of early warning system based on educational big data | |
| Li et al. | Fault identification in power network based on deep reinforcement learning | |
| CN117113166A (en) | Industrial boiler fault detection method based on improved integrated learning | |
| CN105868115A (en) | Building method and system for software test model of software intensive system | |
| Buragohain | Adaptive network based fuzzy inference system (ANFIS) as a tool for system identification with special emphasis on training data minimization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | 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 | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A transformer fault diagnosis method based on AdaBoost algorithm Effective date of registration: 20201119 Granted publication date: 20190806 Pledgee: Xi'an Science and Technology Financial Service Center Co.,Ltd. Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2020980008277 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20211202 Granted publication date: 20190806 Pledgee: Xi'an Science and Technology Financial Service Center Co.,Ltd. Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2020980008277 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A transformer fault diagnosis method based on AdaBoost algorithm Effective date of registration: 20211203 Granted publication date: 20190806 Pledgee: Xi'an Science and Technology Financial Service Center Co.,Ltd. Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2021980014051 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Date of cancellation: 20230109 Granted publication date: 20190806 Pledgee: Xi'an Science and Technology Financial Service Center Co.,Ltd. Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2021980014051 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right | ||
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A Transformer Fault Diagnosis Method Based on Adaboost Algorithm Effective date of registration: 20230309 Granted publication date: 20190806 Pledgee: Industrial Bank Limited by Share Ltd. Xi'an branch Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2023610000156 |
|
| PC01 | Cancellation of the registration of the contract for pledge of patent right | ||
| PC01 | Cancellation of the registration of the contract for pledge of patent right |
Granted publication date: 20190806 Pledgee: Industrial Bank Limited by Share Ltd. Xi'an branch Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2023610000156 |
|
| PE01 | Entry into force of the registration of the contract for pledge of patent right |
Denomination of invention: A Transformer Fault Diagnosis Method Based on Adaboost Algorithm Granted publication date: 20190806 Pledgee: Industrial Bank Limited by Share Ltd. Xi'an branch Pledgor: XI'AN SI-TOP ELECTRIC Co.,Ltd. Registration number: Y2024610000077 |