CN104458923A - Method and system for detecting content warning value of each gas in oil in transformer - Google Patents
Method and system for detecting content warning value of each gas in oil in transformer Download PDFInfo
- Publication number
- CN104458923A CN104458923A CN201410789015.0A CN201410789015A CN104458923A CN 104458923 A CN104458923 A CN 104458923A CN 201410789015 A CN201410789015 A CN 201410789015A CN 104458923 A CN104458923 A CN 104458923A
- Authority
- CN
- China
- Prior art keywords
- transformer
- oil
- dissolved gas
- data
- oil dissolved
- 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 34
- 238000009826 distribution Methods 0.000 claims abstract description 70
- 238000012544 monitoring process Methods 0.000 claims abstract description 28
- 238000004587 chromatography analysis Methods 0.000 claims abstract description 21
- 238000004458 analytical method Methods 0.000 claims description 69
- 238000004519 manufacturing process Methods 0.000 claims description 10
- 238000012417 linear regression Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 abstract 4
- 230000000875 corresponding effect Effects 0.000 abstract 3
- 238000011208 chromatographic data Methods 0.000 abstract 1
- 230000002596 correlated effect Effects 0.000 abstract 1
- 238000010219 correlation analysis Methods 0.000 abstract 1
- 239000002699 waste material Substances 0.000 abstract 1
- 239000007789 gas Substances 0.000 description 136
- 239000004215 Carbon black (E152) Substances 0.000 description 6
- 229930195733 hydrocarbon Natural products 0.000 description 6
- 150000002430 hydrocarbons Chemical class 0.000 description 6
- 230000008569 process Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 3
- HSFWRNGVRCDJHI-UHFFFAOYSA-N alpha-acetylene Natural products C#C HSFWRNGVRCDJHI-UHFFFAOYSA-N 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 125000002534 ethynyl group Chemical group [H]C#C* 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000001257 hydrogen Substances 0.000 description 2
- 229910052739 hydrogen Inorganic materials 0.000 description 2
- 125000004435 hydrogen atom Chemical class [H]* 0.000 description 2
- 239000012774 insulation material Substances 0.000 description 2
- 230000032683 aging Effects 0.000 description 1
- 230000001186 cumulative effect Effects 0.000 description 1
- 238000000354 decomposition reaction Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000004868 gas analysis Methods 0.000 description 1
- 239000007788 liquid Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000036962 time dependent Effects 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Landscapes
- Housings And Mounting Of Transformers (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
Abstract
The invention provides a method for detecting a content warning value of each gas in the oil in a transformer, which includes the realization of the gas in M oil on N transformers. The method comprises the following steps: extracting data, correlated to the calculation of the gas warning value in the oil, in N transformers; training a current calculation model respectively corresponding to the gas content warning value in each oil in N transformers according to the extracted data, wherein the calculation model comprises a statistics distribution model and a correlation analysis model; acquiring a warning value corresponding to the gas content in M oil in the N transformers at the current time according to the current trained calculation model after the training is completed. By implementing the method and the system, factors such as transformer load and weather are comprehensively considered, the oil chromatographic data of each transformer is extracted, the warning value corresponding to each gas content in the oil in each transformer is calculated, the online monitoring warning accuracy and reliability of the transformer oil chromatography can be effectively improved, the resource waste can be avoided, and the equipment risk can be reduced.
Description
Technical field
The present invention relates to power system transformer technical field, particularly relate to and a kind ofly to detect in transformer each gas content in oil and to warn the method and system of value.
Background technology
Dissolved gas analysis is one of most effective method of high-power transformer fault diagnosis.Power industry standard specifies, when transformer under operation, in it during the values listed of oil dissolved gas content more than Fig. 1, should cause and warn, and above-mentioned values listed is also commonly referred to value of warning.
In prior art, above-mentioned value of warning mainly obtains according to 90% cumulative percentage method based on artificial experience, adopt method be: by finding out a collection of typical oil chromatography data (such as having N number of data) from historical data, from big to small sequence after using oil chromatography data individual to (N*90%) as value of warning.But in actual motion, because the operating condition of each transformer is all different, and the impact of the oil chromatography data of each transformer all by himself factor such as load and weather is very large, and therefore each transformer all should formulate the value of warning of self.Along with the continuous expansion of on-line monitoring technique application, the method that tradition obtains based on artificial experience value of warning will cause false alarm and fail to report the phenomenons such as police occurring, and brings the wasting of resources and equipment Risk.
Therefore, need a kind of detection method badly, the value of warning of each gas content in oil in transformer can be obtained.
Summary of the invention
Embodiment of the present invention technical matters to be solved is, there is provided a kind of to detect in transformer each gas content in oil and to warn the method and system of value, by considering the factors such as each transformer load and weather, extract the oil chromatography data of each transformer on oil chromatography on-line monitoring, and to calculate in each transformer each gas content in oil and to warn accordingly value, effectively can improve accuracy and reliability that transformer oil chromatographic on-line monitoring warns, avoid the wasting of resources, reduce equipment Risk.
In order to solve the problems of the technologies described above, embodiments provide and a kind ofly to detect in transformer each gas content in oil and to warn the method for value, it realizes on the N number of transformer including M oil dissolved gas, and described method comprises:
A, extract in described N number of transformer value of all warning with a described M oil dissolved gas and calculate the data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
B, according to described in the data extracted, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
C, warn after the current computation model of value all trained until described M oil dissolved gas content in described N number of transformer, and the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtain the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
Wherein, the concrete steps of described step b comprise:
1 to M in described N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to described first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to described first data and described second data, obtain M*N relation analysis model;
Calculate the degree of fitting of a described M*N statistical distribution pattern and the degree of fitting of a described M*N relation analysis model;
According to the described degree of fitting of the M*N a calculated statistical distribution pattern and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is relation analysis model;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is statistical distribution pattern.
Wherein, the concrete steps of " current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained obtains the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively " in described step c comprising:
When one or more oil dissolved gas content in the described N number of transformer trained warn the current computation model of value be statistical distribution pattern time, obtain by three-sigma principle the value of warning that in N number of transformer described in current time, one or more oil dissolved gas content is corresponding respectively;
When one or more oil dissolved gas content in the described N number of transformer trained warn the current computation model of value be relation analysis model time, the oil temperature threshold value preset by each transformer substitutes in relation analysis model corresponding to described one or more oil dissolved gas respectively, obtains the value of warning that in N number of transformer described in current time, one or more oil dissolved gas content is corresponding respectively.
Wherein, before described step a, described method also comprises:
By chromatography online monitoring system for transformer oil, obtain the first data that described N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that described N number of transformer is corresponding respectively.
Wherein, described first data comprise the gas content of each oil dissolved gas in described N number of transformer and gas production rate corresponding to each oil dissolved gas described; Described second data comprise the accumulated value that oil temperature in described N number of transformer and each oil temperature associated with the duration.
Wherein, described statistical distribution pattern is by carrying out normal distribution matching acquisition to the gas content of current gas; Described relation analysis model is by carrying out linear regression fit acquisition to the gas content of described current gas and the oil temperature of described current gas place transformer.
The embodiment of the present invention additionally provides and a kind ofly to detect in transformer each gas content in oil and to warn the system of value, and it realizes on the N number of transformer including M oil dissolved gas, and described system comprises:
Data extracting unit, calculates for extracting in described N number of transformer value of all warning with a described M oil dissolved gas data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
Model training and determining unit, for the data extracted described in basis, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
Value of warning acquiring unit, after all having trained for the current computation model of value of warning until described M oil dissolved gas content in described N number of transformer, the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtains the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
Wherein, described model training and determining unit comprise:
Model training module, for 1 to M in described N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to described first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to described first data and described second data, obtain M*N relation analysis model;
Degree of fitting computing module, for the degree of fitting of the degree of fitting and a described M*N relation analysis model that calculate a described M*N statistical distribution pattern;
Comparison module, for the degree of fitting of M*N statistical distribution pattern that calculates described in basis and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
First model determination module, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is relation analysis model;
Second model determination module, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is statistical distribution pattern.
Wherein, described system also comprises:
Preset unit, for by chromatography online monitoring system for transformer oil, obtain the first data that described N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that described N number of transformer is corresponding respectively.
Wherein, described first data comprise the gas content of each oil dissolved gas in described N number of transformer and gas production rate corresponding to each oil dissolved gas described; Described second data comprise the accumulated value that oil temperature in described N number of transformer and each oil temperature associated with the duration.
Implement the embodiment of the present invention, there is following beneficial effect:
In embodiments of the present invention, due to the oil chromatography data (gas production rate that gas content and each oil dissolved gas as extracted each each oil dissolved gas of transformer are corresponding) by extracting each transformer on oil chromatography on-line monitoring, and consider the factors such as each transformer load and weather (as the accumulated value that the oil temperature in each transformer that extraction transformer temperature monitoring system gathers and each oil temperature associated with the duration), train the computation model of degree of fitting the best, and according to this optimal computed model, obtain the value of warning that in each transformer of current time, each oil dissolved gas content is corresponding respectively, effectively can improve accuracy and reliability that transformer oil chromatographic on-line monitoring warns, avoid the wasting of resources, reduce equipment Risk.
Accompanying drawing explanation
In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, the accompanying drawing obtaining other according to these accompanying drawings still belongs to category of the present invention.
Fig. 1 is that the oil dissolved gas content of transformer in prior art is warned the schematic diagram of value;
In the detection transformer that Fig. 2 provides for the embodiment of the present invention, in oil, each gas content is warned the process flow diagram of method of value;
In the detection transformer that Fig. 3 provides for the embodiment of the present invention, in oil, each gas content is warned the structural representation of system of value.
Embodiment
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, the present invention is described in further detail.
Inventor finds, when the paper oil insulation material of inside transformer normally runs, and meeting aging Sum decomposition gradually under heat and electric effect, thus produce a small amount of H
2, low molecular hydrocarbon gas, CO, CO
2deng gas, and equally when heat and electric fault, the paper oil insulation material of inside transformer also can produce above-mentioned gas, the source of above-mentioned gas cannot be distinguished technically, the numerical value of its correspondence does not have strict boundary yet, and the source of above-mentioned gas and the load of transformer and temperature, O in oil
2many variable factors such as content and liquid water content, oily protection system and the circulation system are relevant.
Because the factor such as load and temperature of transformer is always in time in continuous change, therefore the oil dissolved gas content of transformer is warned value also corresponding time to time change, but the relation of the factor such as load, temperature of oil dissolved gas content and transformer in the industry cycle not yet has unified qualitative understanding at present, actual production run cannot be instructed by quantitative model.
Therefore, inventor's knowledge based finds principle, proposes a kind ofly to detect in transformer each gas content in oil and to warn the method for value, in order to determine that the time dependent oil dissolved gas content of each transformer is warned value.As shown in Figure 2, what provide for the present inventor a kind ofly to detect in transformer each gas content in oil and to warn the method for value, and it realizes on the N number of transformer including M oil dissolved gas, and described method comprises:
Step S101, extract in described N number of transformer value of all warning with a described M oil dissolved gas and calculate the data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
Detailed process is, before step S101, described method also comprises: by chromatography online monitoring system for transformer oil, obtains the first data that N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that N number of transformer is corresponding respectively.
In the embodiment of the present invention, in order to can in the acquisition transformer of precision in oil each gas content to warn value, in the process of analytical algorithm, namely the data extracted consider the oil chromatography data (the first data as being associated with M oil dissolved gas in each transformer) of each transformer on oil chromatography on-line monitoring, have also contemplated that the factors such as each transformer load and weather and the oil temperature delta data (the second data as being associated with oil temperature in each transformer) that produces.Wherein, first data comprise the data directly got from chromatography online monitoring system for transformer oil, as the gas content data etc. of hydrogen, acetylene, total hydrocarbon, and to the data that oil chromatography online monitoring data is processed, as the gas production rate etc. of hydrogen, acetylene, total hydrocarbon, therefore the first data are defined as the data of the gas content that comprises each oil dissolved gas in N number of transformer and gas production rate corresponding to each oil dissolved gas etc.; Second data comprise the temperature of oil in transformer directly obtained from transformer temperature monitoring system and the data of processing temperature of oil in transformer, as oil temperature and the accumulated value etc. of duration, therefore the second data are defined as the data comprising accumulated value that oil temperature in N number of transformer and each oil temperature associate with the duration etc.
It should be noted that the data such as oil chromatography data and oil temperature of acquisition, all for the data-interface by having set up with all kinds of on-line monitoring system and the online monitoring data got.
Step S102, according to described in the data extracted, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
Detailed process is, 1 to M in N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to the first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to the first data and the second data, obtain M*N relation analysis model;
Calculate the degree of fitting of M*N statistical distribution pattern and the degree of fitting of M*N relation analysis model;
According to the degree of fitting of M*N the statistical distribution pattern calculated and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then determine that the warn current computation model of value of one or more oil dissolved gas content in N number of transformer of detecting is relation analysis model; Wherein, relation analysis model is by carrying out linear regression fit acquisition to the gas content of described current gas and the oil temperature of described current gas place transformer;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then determine that the warn current computation model of value of one or more oil dissolved gas content in N number of transformer of detecting is statistical distribution pattern; Wherein, statistical distribution pattern is by carrying out normal distribution matching acquisition to the gas content of current gas.
As an example, the application scenarios of step S102 is further illustrated:
(1) start to train the warn computation model of value of a jth oil dissolved gas (as total hydrocarbon) of i-th transformer;
(2) corresponding statistical distribution pattern is trained, as carried out normal distribution matching to the gas content data of total hydrocarbon according to the first data; Corresponding relation analysis model is jointly trained, as carried out linear regression fit to total hydrocarbon content and oil temperature data according to the first data and the second data;
(3) calculate the degree of fitting of statistical distribution pattern corresponding to a jth oil dissolved gas and the relation analysis model degree of fitting of correspondence thereof respectively, select the good model of degree of fitting as optimization model;
(4) when the degree of fitting of statistical distribution pattern corresponding to a jth oil dissolved gas is greater than and the relation analysis model degree of fitting of correspondence time, the warn computation model of value of the jth oil dissolved gas of this i-th transformer is the statistical distribution pattern that a jth oil dissolved gas is corresponding;
(5) when the degree of fitting of statistical distribution pattern corresponding to a jth oil dissolved gas is less than and the relation analysis model degree of fitting of correspondence time, the warn computation model of value of the jth oil dissolved gas of this i-th transformer is the relation analysis model model that a jth oil dissolved gas is corresponding.
Step S103, warn after the current computation model of value all trained until described M oil dissolved gas content in described N number of transformer, and the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtain the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
Detailed process is, after all computation models have been trained, when oil dissolved gas content one or more in the N number of transformer trained warn the current computation model of value be statistical distribution pattern time, obtain by three-sigma principle the value of warning that in the N number of transformer of current time, one or more oil dissolved gas content is corresponding respectively;
When oil dissolved gas content one or more in the N number of transformer trained warn the current computation model of value be relation analysis model time, the oil temperature threshold value preset by each transformer substitutes in relation analysis model corresponding to one or more oil dissolved gas respectively, obtains the value of warning that in the N number of transformer of current time, one or more oil dissolved gas content is corresponding respectively.
It should be noted that the oil temperature threshold value that the relation analysis model of current one or more oil dissolved gas substitutes into is corresponding with default oil temperature threshold value in its place transformer.
As shown in Figure 3, the embodiment of the present invention additionally provides and a kind ofly to detect in transformer each gas content in oil and to warn the system of value, and it realizes on the N number of transformer including M oil dissolved gas, and described system comprises:
Data extracting unit 110, calculates for extracting in described N number of transformer value of all warning with a described M oil dissolved gas data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
Model training and determining unit 120, for the data extracted described in basis, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
Value of warning acquiring unit 130, after all having trained for the current computation model of value of warning until described M oil dissolved gas content in described N number of transformer, the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtains the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
Wherein, described model training and determining unit 120 comprise:
Model training module 1201, for 1 to M in described N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to described first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to described first data and described second data, obtain M*N relation analysis model;
Degree of fitting computing module 1202, for the degree of fitting of the degree of fitting and a described M*N relation analysis model that calculate a described M*N statistical distribution pattern;
Comparison module 1203, for the degree of fitting of M*N statistical distribution pattern that calculates described in basis and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
First model determination module 1204, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is relation analysis model;
Second model determination module 1205, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is statistical distribution pattern.
Wherein, described system also comprises:
Preset unit 140, for by chromatography online monitoring system for transformer oil, obtain the first data that described N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that described N number of transformer is corresponding respectively.
Wherein, described first data comprise the gas content of each oil dissolved gas in described N number of transformer and gas production rate corresponding to each oil dissolved gas described; Described second data comprise the accumulated value that oil temperature in described N number of transformer and each oil temperature associated with the duration.
Implement the embodiment of the present invention, there is following beneficial effect:
In embodiments of the present invention, due to the oil chromatography data (gas production rate that gas content and each oil dissolved gas as extracted each each oil dissolved gas of transformer are corresponding) by extracting each transformer on oil chromatography on-line monitoring, and consider the factors such as each transformer load and weather (as the accumulated value that the oil temperature in each transformer that extraction transformer temperature monitoring system gathers and each oil temperature associated with the duration), train the computation model of degree of fitting the best, and according to this optimal computed model, obtain the value of warning that in each transformer of current time, each oil dissolved gas content is corresponding respectively, effectively can improve accuracy and reliability that transformer oil chromatographic on-line monitoring warns, avoid the wasting of resources, reduce equipment Risk.
It should be noted that in said system embodiment, each included system unit is carry out dividing according to function logic, but is not limited to above-mentioned division, as long as can realize corresponding function; In addition, the concrete title of each functional unit, also just for the ease of mutual differentiation, is not limited to protection scope of the present invention.
One of ordinary skill in the art will appreciate that all or part of step realized in above-described embodiment method is that the hardware that can carry out instruction relevant by program has come, described program can be stored in a computer read/write memory medium, described storage medium, as ROM/RAM, disk, CD etc.
Above disclosedly be only a kind of preferred embodiment of the present invention, certainly can not limit the interest field of the present invention with this, therefore according to the equivalent variations that the claims in the present invention are done, still belong to the scope that the present invention is contained.
Claims (10)
1. to detect in transformer each gas content in oil to warn a method for value, it is characterized in that, it realizes on the N number of transformer including M oil dissolved gas, and described method comprises:
A, extract in described N number of transformer value of all warning with a described M oil dissolved gas and calculate the data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
B, according to described in the data extracted, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
C, warn after the current computation model of value all trained until described M oil dissolved gas content in described N number of transformer, and the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtain the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
2. the method for claim 1, is characterized in that, the concrete steps of described step b comprise:
1 to M in described N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to described first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to described first data and described second data, obtain M*N relation analysis model;
Calculate the degree of fitting of a described M*N statistical distribution pattern and the degree of fitting of a described M*N relation analysis model;
According to the described degree of fitting of the M*N a calculated statistical distribution pattern and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is relation analysis model;
When detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is statistical distribution pattern.
3. the method for claim 1, it is characterized in that, the concrete steps of " current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained obtains the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively " in described step c comprising:
When one or more oil dissolved gas content in the described N number of transformer trained warn the current computation model of value be statistical distribution pattern time, obtain by three-sigma principle the value of warning that in N number of transformer described in current time, one or more oil dissolved gas content is corresponding respectively;
When one or more oil dissolved gas content in the described N number of transformer trained warn the current computation model of value be relation analysis model time, the oil temperature threshold value preset by each transformer substitutes in relation analysis model corresponding to described one or more oil dissolved gas respectively, obtains the value of warning that in N number of transformer described in current time, one or more oil dissolved gas content is corresponding respectively.
4. the method for claim 1, is characterized in that, before described step a, described method also comprises:
By chromatography online monitoring system for transformer oil, obtain the first data that described N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that described N number of transformer is corresponding respectively.
5. the method according to any one of Claims 1-4, is characterized in that, described first data comprise the gas content of each oil dissolved gas in described N number of transformer and gas production rate corresponding to each oil dissolved gas described; Described second data comprise the accumulated value that oil temperature in described N number of transformer and each oil temperature associated with the duration.
6. method as claimed in claim 5, it is characterized in that, described statistical distribution pattern is by carrying out normal distribution matching acquisition to the gas content of current gas; Described relation analysis model is by carrying out linear regression fit acquisition to the gas content of described current gas and the oil temperature of described current gas place transformer.
7. to detect in transformer each gas content in oil to warn a system for value, it is characterized in that, it realizes on the N number of transformer including M oil dissolved gas, and described system comprises:
Data extracting unit, calculates for extracting in described N number of transformer value of all warning with a described M oil dissolved gas data be associated; Wherein, described data comprise the first data be associated with a described M oil dissolved gas in each transformer, and the second data be associated with oil temperature in each transformer; M, N are positive integer;
Model training and determining unit, for the data extracted described in basis, train each oil dissolved gas content in N number of transformer and to warn the current computation model of value correspondence respectively; Wherein, described computation model comprises the statistical distribution pattern trained according to described first data and the relation analysis model jointly trained according to described first data and described second data;
Value of warning acquiring unit, after all having trained for the current computation model of value of warning until described M oil dissolved gas content in described N number of transformer, the current computation model of value of warning according to each oil dissolved gas content in the described N number of transformer trained, obtains the value of warning that in N number of transformer described in current time, described M oil dissolved gas content is corresponding respectively.
8. system as claimed in claim 7, it is characterized in that, described model training and determining unit comprise:
Model training module, for 1 to M in described N number of transformer oil dissolved gas is all trained corresponding statistical distribution pattern according to described first data, obtain M*N statistical distribution pattern, and all jointly train corresponding relation analysis model according to described first data and described second data, obtain M*N relation analysis model;
Degree of fitting computing module, for the degree of fitting of the degree of fitting and a described M*N relation analysis model that calculate a described M*N statistical distribution pattern;
Comparison module, for the degree of fitting of M*N statistical distribution pattern that calculates described in basis and the degree of fitting of M*N relation analysis model, the size cases of the degree of fitting of the relation analysis model that the degree of fitting detecting the statistical distribution pattern that same oil dissolved gas obtains in same transformer successively in 1 to N number of transformer obtains with it;
First model determination module, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all less than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is relation analysis model;
Second model determination module, for when detecting that the degree of fitting of the statistical distribution pattern that one or more oil dissolved gas obtains in described N number of transformer is all greater than the degree of fitting of the relation analysis model that it obtains, then in the N number of transformer detected described in determining, the warn current computation model of value of one or more oil dissolved gas content is statistical distribution pattern.
9. system as claimed in claim 7, it is characterized in that, described system also comprises:
Preset unit, for by chromatography online monitoring system for transformer oil, obtain the first data that described N number of transformer is corresponding respectively, and by transformer temperature monitoring system, obtain the second data that described N number of transformer is corresponding respectively.
10. the system according to any one of claim 7 to 9, is characterized in that, described first data comprise the gas content of each oil dissolved gas in described N number of transformer and gas production rate corresponding to each oil dissolved gas described; Described second data comprise the accumulated value that oil temperature in described N number of transformer and each oil temperature associated with the duration.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410789015.0A CN104458923B (en) | 2014-12-19 | 2014-12-19 | A kind ofly to detect in transformer each gas content in oil and to warn the method and system of value |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410789015.0A CN104458923B (en) | 2014-12-19 | 2014-12-19 | A kind ofly to detect in transformer each gas content in oil and to warn the method and system of value |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104458923A true CN104458923A (en) | 2015-03-25 |
| CN104458923B CN104458923B (en) | 2016-04-13 |
Family
ID=52905316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410789015.0A Active CN104458923B (en) | 2014-12-19 | 2014-12-19 | A kind ofly to detect in transformer each gas content in oil and to warn the method and system of value |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104458923B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104792467A (en) * | 2015-04-09 | 2015-07-22 | 贵阳供电局 | Method for judging internal leakage possibility of on-load tap-changer according to characteristic gas content in oil |
| CN107664690A (en) * | 2017-07-28 | 2018-02-06 | 国网信通亿力科技有限责任公司 | A kind of method for predicting gas dissolved in oil of power trans-formers |
| CN113252827A (en) * | 2021-05-12 | 2021-08-13 | 国网安徽省电力有限公司电力科学研究院 | Transformer oil chromatographic device performance evaluation method and system based on chi-square test |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1292491A (en) * | 1999-09-17 | 2001-04-25 | 通用电气公司 | Fluid-filled electric equipment intelligent analytical system and method |
| CN104063577A (en) * | 2014-05-12 | 2014-09-24 | 吉林省电力科学研究院有限公司 | Method for forecasting characteristic gas development tendency in transformer oil based on generalized recurrent neural network |
-
2014
- 2014-12-19 CN CN201410789015.0A patent/CN104458923B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1292491A (en) * | 1999-09-17 | 2001-04-25 | 通用电气公司 | Fluid-filled electric equipment intelligent analytical system and method |
| CN104063577A (en) * | 2014-05-12 | 2014-09-24 | 吉林省电力科学研究院有限公司 | Method for forecasting characteristic gas development tendency in transformer oil based on generalized recurrent neural network |
Non-Patent Citations (3)
| Title |
|---|
| W. A. FESSLER 等: "A REFINED MATHEMATICAL MODEL FOR PREDICTION OF BUBBLE EVOLUTION IN TRANSFORMERS", 《POWER DELIVERY》 * |
| 姚森敬 等: "一种变压器状态评估中的状态量优选方法", 《中国电力》 * |
| 梁永亮 等: "变压器油色谱在线监测周期动态调整策略研究", 《中国电机工程学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104792467A (en) * | 2015-04-09 | 2015-07-22 | 贵阳供电局 | Method for judging internal leakage possibility of on-load tap-changer according to characteristic gas content in oil |
| CN104792467B (en) * | 2015-04-09 | 2017-10-13 | 贵阳供电局 | It is a kind of to judge to leak the method for possibility in load ratio bridging switch based on characteristic gas content in oil |
| CN107664690A (en) * | 2017-07-28 | 2018-02-06 | 国网信通亿力科技有限责任公司 | A kind of method for predicting gas dissolved in oil of power trans-formers |
| CN113252827A (en) * | 2021-05-12 | 2021-08-13 | 国网安徽省电力有限公司电力科学研究院 | Transformer oil chromatographic device performance evaluation method and system based on chi-square test |
| CN113252827B (en) * | 2021-05-12 | 2022-07-22 | 国网安徽省电力有限公司电力科学研究院 | Method and system for evaluating performance of transformer oil chromatographic device based on chi-square test |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104458923B (en) | 2016-04-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106888205B (en) | Non-invasive PLC anomaly detection method based on power consumption analysis | |
| CN105279365B (en) | For the method for the sample for learning abnormality detection | |
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| CN103903408B (en) | Method for early warning and system are investigated in equipment fault | |
| CN109739904B (en) | Time sequence marking method, device, equipment and storage medium | |
| CN109470946B (en) | Power generation equipment fault detection method and system | |
| CN112149860A (en) | Automatic anomaly detection method and system | |
| CN104458923B (en) | A kind ofly to detect in transformer each gas content in oil and to warn the method and system of value | |
| CN111523683B (en) | Method and system for predicting technological parameters in tobacco processing | |
| CN103103570B (en) | Based on the aluminium cell condition diagnostic method of pivot similarity measure | |
| CN113742993B (en) | Method, device, equipment and storage medium for predicting life loss of dry-type transformer | |
| CN104635724A (en) | Abnormity detection method for natural gas purification process based on analysis of independent component of dynamic kernel | |
| CN103699514A (en) | Method for detecting steady state and judging operation condition of water treatment process in thermal power plant | |
| CN113084388B (en) | Welding quality detection method, system, device and storage medium | |
| CN112183835B (en) | Water guide tile temperature trend early warning method, device and system based on machine learning | |
| CN112036581A (en) | Performance detection method and device of vehicle air conditioning system, storage medium and terminal | |
| CN105425739A (en) | System for predicting abnormality occurrence using PLC log data | |
| CN103529337B (en) | The recognition methods of nonlinear correlation relation between equipment failure and electric quantity information | |
| CN117520951B (en) | Transformer health assessment method and system based on multiple characteristic quantities | |
| CN109142979B (en) | Method and device for detecting abnormal state of power distribution network | |
| CN104021232A (en) | Configuration item-based aircraft test data interpretation method | |
| CN117250449A (en) | Transformer insulation state evaluation method and device, electronic equipment and storage medium | |
| CN112241751B (en) | Anomaly detection method and system, and computer readable storage medium | |
| WO2023125586A1 (en) | Training method and apparatus for urban underground gas leakage identification model | |
| CN111181172B (en) | Power grid frequency disturbance source positioning method for scheduling master station |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CP03 | Change of name, title or address | ||
| CP03 | Change of name, title or address |
Address after: 518000 electric power dispatching and communication building, 4020 Shennan East Road, Luohu District, Shenzhen, Guangdong Patentee after: Shenzhen Power Supply Co.,Ltd. Country or region after: China Patentee after: China Southern Power Grid Digital Platform Technology (Guangdong) Co.,Ltd. Address before: 518000 electric power dispatching and communication building, 4020 Shennan East Road, Luohu District, Shenzhen, Guangdong Patentee before: Shenzhen Power Supply Co.,Ltd. Country or region before: China Patentee before: SHENZHEN COMTOP INFORMATION TECHNOLOGY Co.,Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20240403 Address after: 518000 electric power dispatching and communication building, 4020 Shennan East Road, Luohu District, Shenzhen, Guangdong Patentee after: Shenzhen Power Supply Co.,Ltd. Country or region after: China Address before: 518000 electric power dispatching and communication building, 4020 Shennan East Road, Luohu District, Shenzhen, Guangdong Patentee before: Shenzhen Power Supply Co.,Ltd. Country or region before: China Patentee before: China Southern Power Grid Digital Platform Technology (Guangdong) Co.,Ltd. |