CN107664690B - A method of prediction gas dissolved in oil of power trans-formers - Google Patents
A method of prediction gas dissolved in oil of power trans-formers Download PDFInfo
- Publication number
- CN107664690B CN107664690B CN201710660760.9A CN201710660760A CN107664690B CN 107664690 B CN107664690 B CN 107664690B CN 201710660760 A CN201710660760 A CN 201710660760A CN 107664690 B CN107664690 B CN 107664690B
- Authority
- CN
- China
- Prior art keywords
- oil
- concentration
- transformer
- data
- gas
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000012545 processing Methods 0.000 claims abstract description 29
- 238000010219 correlation analysis Methods 0.000 claims abstract description 4
- 239000007789 gas Substances 0.000 claims description 86
- VNWKTOKETHGBQD-UHFFFAOYSA-N methane Chemical compound C VNWKTOKETHGBQD-UHFFFAOYSA-N 0.000 claims description 24
- HSFWRNGVRCDJHI-UHFFFAOYSA-N Acetylene Chemical compound C#C HSFWRNGVRCDJHI-UHFFFAOYSA-N 0.000 claims description 16
- 125000002534 ethynyl group Chemical group [H]C#C* 0.000 claims description 14
- 239000001257 hydrogen Substances 0.000 claims description 12
- 229910052739 hydrogen Inorganic materials 0.000 claims description 12
- CURLTUGMZLYLDI-UHFFFAOYSA-N Carbon dioxide Chemical compound O=C=O CURLTUGMZLYLDI-UHFFFAOYSA-N 0.000 claims description 10
- VGGSQFUCUMXWEO-UHFFFAOYSA-N Ethene Chemical compound C=C VGGSQFUCUMXWEO-UHFFFAOYSA-N 0.000 claims description 10
- 239000005977 Ethylene Substances 0.000 claims description 10
- OTMSDBZUPAUEDD-UHFFFAOYSA-N Ethane Chemical compound CC OTMSDBZUPAUEDD-UHFFFAOYSA-N 0.000 claims description 9
- UFHFLCQGNIYNRP-UHFFFAOYSA-N Hydrogen Chemical compound [H][H] UFHFLCQGNIYNRP-UHFFFAOYSA-N 0.000 claims description 8
- UGFAIRIUMAVXCW-UHFFFAOYSA-N Carbon monoxide Chemical compound [O+]#[C-] UGFAIRIUMAVXCW-UHFFFAOYSA-N 0.000 claims description 6
- 229910002092 carbon dioxide Inorganic materials 0.000 claims description 6
- 229910002091 carbon monoxide Inorganic materials 0.000 claims description 6
- 229910052799 carbon Inorganic materials 0.000 claims description 5
- 239000001569 carbon dioxide Substances 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 238000011021 bench scale process Methods 0.000 claims description 2
- GWEVSGVZZGPLCZ-UHFFFAOYSA-N Titan oxide Chemical compound O=[Ti]=O GWEVSGVZZGPLCZ-UHFFFAOYSA-N 0.000 claims 2
- OKTJSMMVPCPJKN-UHFFFAOYSA-N Carbon Chemical compound [C] OKTJSMMVPCPJKN-UHFFFAOYSA-N 0.000 claims 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 claims 1
- 230000015572 biosynthetic process Effects 0.000 claims 1
- 239000001301 oxygen Substances 0.000 claims 1
- 229910052760 oxygen Inorganic materials 0.000 claims 1
- 239000004408 titanium dioxide Substances 0.000 claims 1
- 238000007405 data analysis Methods 0.000 description 9
- 150000002431 hydrogen Chemical class 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 240000002853 Nelumbo nucifera Species 0.000 description 1
- 235000006508 Nelumbo nucifera Nutrition 0.000 description 1
- 235000006510 Nelumbo pentapetala Nutrition 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000032683 aging Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 230000004888 barrier function Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000002596 correlated effect Effects 0.000 description 1
- 230000000875 corresponding effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000006866 deterioration Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000013399 early diagnosis Methods 0.000 description 1
- 238000013210 evaluation model Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000004868 gas analysis Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N33/00—Investigating or analysing materials by specific methods not covered by groups G01N1/00 - G01N31/00
- G01N33/26—Oils; Viscous liquids; Paints; Inks
- G01N33/28—Oils, i.e. hydrocarbon liquids
- G01N33/2835—Specific substances contained in the oils or fuels
- G01N33/2841—Gas in oils, e.g. hydrogen in insulating oils
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Data Mining & Analysis (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Computational Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Physics (AREA)
- Economics (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Immunology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Pathology (AREA)
- Medicinal Chemistry (AREA)
- Evolutionary Biology (AREA)
- Food Science & Technology (AREA)
- Oil, Petroleum & Natural Gas (AREA)
- Operations Research (AREA)
- Probability & Statistics with Applications (AREA)
- General Chemical & Material Sciences (AREA)
- Algebra (AREA)
- Chemical Kinetics & Catalysis (AREA)
- Databases & Information Systems (AREA)
Abstract
The present invention provides a kind of predicting model for dissolved gas in transformer oil methods, including:Obtain each characteristic gas history concentration data of transformer, history oil temperature and historical load data;Dimensionless processing is carried out to each characteristic gas historical data of the transformer, history oil temperature and historical load;Carry out the correlation analysis between data;Build transformer equipment oil dissolved gas concentration prediction model;The transformer equipment oil dissolved gas concentration prediction model is optimized;It predicts dissolving the single features gas concentration in transformer oil.The method of the present invention is capable of the variation tendency of Accurate Prediction oil soluble gas bulk concentration, effectively identifies the Hidden fault of transformer equipment, realizes the controls in advance to equipment.
Description
Technical field
The present invention relates to field of electric power automation, more particularly to a kind of predicting model for dissolved gas in transformer oil method.
Background technology
The best approach of transformer fault early diagnosis is mainly dissolved gas analysis method, by molten in transformer oil
Solution characteristic gas concentration can capture the omen of transformer equipment failure.In practice, typical transformer oil dissolved gas point
Analysis method monitors and judges operating condition of the transformer when oil sample acquires, find in time the hiding defect of transformer or therefore
Barrier.Its principle is:Oil-filled power transformer, by effect meeting aging and deterioration electrically or thermally, generates in long-term operational process
A small amount of gas (commonly referred to as characteristic gas) is dissolved in transformer oil.Generally speaking the amount of dissolved gas is one and gradually puts down
The process surely accumulated, but in failure or abnormal generation, gas content numerical value can have greatly changed.
Consider the influence of the aerogenesis principle and temperature of oil in transformer, load of characteristic of transformer gas to characteristic gas,
Using big data analysis algorithm, explore interrelated relationship in transformer oil between dissolving characteristic gas and characteristic gas with
Incidence relation between transformer equipment load, top-oil temperature excavates the factor larger with object gas correlation to be predicted
As input variable, the variation of Accurate Prediction oil soluble gas in future bulk concentration effectively identifies the Hidden fault of transformer equipment, real
Now to the controls in advance of equipment.
Invention content
The present invention to solve the above problems, provide it is a kind of prediction gas dissolved in oil of power trans-formers method, wherein
The transformer oil is the transformer equipment oil in work, which is characterized in that the method includes:
Step 000:Characteristic gas is selected in the dissolved gas of the transformer oil, measures the concentration of the characteristic gas
And the oil temperature and load of the transformer are recorded simultaneously;
Step 100:Obtain the history oil temperature data and historical load data of the transformer;
Step 200:Obtain the history concentration data of the characteristic gas;
Step 300:The history concentration data, history oil temperature data and historical load data are carried out at nondimensionalization
Reason;
Step 400:Correlation and the characteristic gas between the characteristic gas and the oil temperature and described negative
Correlation between lotus is analyzed;
Step 500:The correlation determined according to step 400 builds the prediction mould of the gas dissolved in oil of power trans-formers
Type;
Step 600:The prediction model of the gas dissolved in oil of power trans-formers is optimized;
Step 700:It selects the concentration of one of described characteristic gas as output variable, selects the concentration of other characteristic gas
And the oil temperature and the load be as input variable, using the prediction model of the gas dissolved in oil of power trans-formers,
The concentration of one of the characteristic gas is predicted.
The method of the present invention is capable of the variation tendency of Accurate Prediction oil soluble gas bulk concentration, effectively identifies the latent of transformer equipment
Volt property failure, realizes to the controls in advance of equipment, reduces the failure rate of transformer, while saving manpower and materials, improve
The social image of national grid brings huge economic benefit and social benefit.
Description of the drawings
Fig. 1 is the flow diagram of the present invention.
Specific implementation mode
It to make the object, technical solutions and advantages of the present invention clearer, below will be to the technology in the embodiment of the present invention
Scheme is clearly and completely described, it is clear that and described embodiments are some of the embodiments of the present invention, rather than whole
Embodiment.Based on the embodiments of the present invention, those of ordinary skill in the art are obtained without creative efforts
The every other embodiment obtained, shall fall within the protection scope of the present invention.
In the description of the present invention, it is to be understood that, term " longitudinal direction ", " transverse direction ", "upper", "lower", "front", "rear",
The orientation or positional relationship of the instructions such as "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outside" is based on attached drawing institute
The orientation or positional relationship shown is merely for convenience of the description present invention, does not indicate or imply the indicated device or element must
There must be specific orientation, with specific azimuth configuration and operation, therefore be not considered as limiting the invention.
As shown in Fig. 1, the embodiment of the present invention one discloses a kind of predicting model for dissolved gas in transformer oil method,
It is characterized in that, including:
Step 100:Obtain transformer history oil temperature and historical load data;
Transformer history oil temperature is obtained in step 100, the method for historical load data is:From electric power OMS system (power grid tune
Spend operation management system) in extract the history oil temperature and historical load data of prediction transformer equipment on the 30th a few days ago, and count
To 95 probability value of maximum value and historical load data of the history oil temperature of each main transformer;
Step 200:Obtain each characteristic gas history concentration data of the transformer;
The method that each characteristic gas history concentration data of the transformer is obtained in step 200 is:From electric power OMS system
Each characteristic gas history concentration data for the transformer for predicting 30 days a few days ago is extracted in (dispatching of power netwoks operation management system):
Step 300:Each characteristic gas historical data of the transformer, history oil temperature and historical load are carried out at dimensionless
Reason;
The step 300 further comprises:
Step 310:To H2(hydrogen) concentration carries out dimensionless processing:
Investigate the historical data in hydrogen 30 days, the results showed that, density of hydrogen 30 days mean values one third to three times
It is fluctuated in section, stability is poor, therefore carries out dimensionless processing to density of hydrogen using following methods, it is ensured that hydrogen
99.8% or more concentration value falls into interval;
WhereinFor density of hydrogen in i-th day transformer oil,ForAs a result, n is the change after nondimensionalization
Density of hydrogen statistics number of days in depressor oil, preferably 30 days;
Step 320:To CH4(methane) concentration carries out dimensionless processing:
Investigate the historical data in methane 30 days, the results showed that, methane concentration minimum value and maximum value magnitude difference are larger,
Maximum value may be tens of thousands of times of minimum value or more, therefore carry out dimensionless processing to methane concentration using following methods, can be with
Ensure that treated that methane concentration value discreteness is unlikely to excessive;
WhereinFor methane concentration in i-th day transformer oil,ForAfter nondimensionalization as a result,For
The maximum value of methane concentration in the transformer oil,Expression rounds up;
Step 330:To C2H6(ethane) concentration carries out dimensionless processing:
Investigate the historical data in acetylene 30 days, the results showed that, concentration of acetylene is substantially in the 80%-120% of 30 days mean values
It is fluctuated in section, stability is good, therefore carries out dimensionless processing to concentration of acetylene using following methods, can simplify calculating, carry
High efficiency;
WhereinFor ethane concentration in i-th day transformer oil,ForAfter nondimensionalization as a result,The maximum value and minimum value of ethane concentration in the respectively described transformer oil;
Step 340:To C2H4(ethylene) concentration carries out dimensionless processing:
Investigate the historical data in ethylene 30 days, the results showed that, ethylene concentration minimum value shows weight close to 0
Tail is distributed, therefore carries out dimensionless processing to ethylene concentration using following methods, can improve the precision of data processing;
WhereinFor ethylene concentration in i-th day transformer oil,For ethylene in i-th day transformer oil
ConcentrationAfter nondimensionalization as a result,Interval be
Step 350:To C2H2(acetylene) concentration carries out dimensionless processing:
Investigate the historical data in acetylene 30 days, the results showed that, concentration of acetylene fluctuation is larger, but it is big show mean value centered on
It is bell-like symmetrical, therefore dimensionless processing is carried out to concentration of acetylene using following methods, it is dense can preferably to weigh acetylene
The irrelevance of degree and mean value;
WhereinFor concentration of acetylene in i-th day transformer oil,For acetylene in i-th day transformer oil
ConcentrationResult after nondimensionalization;
Step 360:Dimensionless processing is carried out to CO (carbon monoxide) concentration:
Investigate the historical data in carbon monoxide 30 days, the results showed that, the distribution of carbonomonoxide concentration appears similar to
Exponential distribution, have it is without memory, but therefore use following methods to carbonomonoxide concentration carry out dimensionless processing so that it is immeasurable
Guiding principle treated carbon monoxide distribution is more uniform;
WhereinFor carbonomonoxide concentration in i-th day transformer oil,It is in i-th day transformer oil one
Aoxidize concentration of carbonAfter nondimensionalization as a result,Indicate the nature truth of a matterPower;
Step 370:Dimensionless processing is carried out to CO2 (carbon dioxide) concentration:
Investigate the historical data in carbon dioxide 30 days, the results showed that, the distribution of gas concentration lwevel shows chaos spy
Property, having no relevance on space-time can say, therefore carry out dimensionless processing to gas concentration lwevel using following methods so that immeasurable
Guiding principle treated gas concentration lwevel has certain regularity;
WhereinFor gas concentration lwevel in i-th day transformer oil,It is in i-th day transformer oil two
Aoxidize concentration of carbonAs a result, m is User Defined constant after nondimensionalization, and m ∈ [1, n);Bench-scale testing the result shows that:
When n is preferably 30 days, if m values are 17, nondimensionalization best results;
Step 380:Dimensionless processing is carried out to history oil temperature:
History oil temperature is more stable, therefore can simplify processing step, improves treatment effeciency;
Wherein TEMi' be i-th day transformer history oil temperature, TEMi' be i-th day transformer history oil temperature
TEMiAs a result, TEM after nondimensionalizationmax、TEMminThe maximum value and minimum value of the respectively described transformer oil history oil temperature;
Step 390:Dimensionless processing is carried out to historical load:
There is transformer rated load, historical load to deviate less, so using following processing method:
Wherein ViFor i-th day transformer historical load data, Vi' it is i-th day transformer historical load data Vi
As a result, V after nondimensionalizationstdFor the rated load of the transformer;
Step 400:It carries out between each characteristic gas of the transformer and each characteristic gas of the transformer and the oil
Correlation analysis between warm, the described load forms correlation crosstab.Each list item in correlation crosstab is first
Correlation between data and the second data, wherein the first data are by step 300 treated various features gas concentration
In one, the second data be by one in step 300 treated various features gas concentration or oil temperature, load,
First data are different from the second data.Such as first data be carbon monoxide concentration, the second data be carbon dioxide concentration,
For another example, the first data are the concentration of hydrogen, and the second data are oil temperature etc..
First data are { FST }, { FST }={ FST (t1),FST(t2),...,FST(tp), { t1,t2…tpIt is to fix
The time point sequence of time interval, time interval are denoted as Δ t, time point quantity p, tiFor i-th of time point, i ∈ [1, p],
FST(ti) it is value of first data i-th of time point, the second data set is { SCD }, { SCD }={ SCD (t1),SCD
(t2),...,SCD(tp), the first data are calculated with the second data in the correlation at i-th of time point, are denoted as
Wherein Δ oTiFor the difference of i-th time point the first data and the second data, Δ minTFor the first data and second
The minimum value of the difference of data, Δ maxTFor the maximum value of the first data and the difference of the second data, ρTFor the first data and second
Resolution ratio between data, ρT∈ [0,1], preferably 0.5;
Calculate the correlation r between the first data and the second dataC,
If rTMore than or equal to first threshold, then it is assumed that it is related between the first data and the second data, if rTLess than first
Threshold value, then it is assumed that uncorrelated between the first data and the second data;By testing on a small scale, first threshold is preferably 0.8;
Because the first data are by one in step 300 treated various features gas concentration, the second data are
By one in step 300 treated various features gas concentration or oil temperature, load, the first data are different from the second number
According to, it is possible to it calculates between each characteristic gas and the relationship between characteristic gas and history oil temperature, historical load, shape
At correlation crosstab;The correlation crosstab at most may includeKind correlative relationship;
Compare the correlation between each characteristic gas and between oil temperature, load in crosstab, correlation is bigger, explanation
Correlation degree between each characteristic gas and between characteristic gas and oil temperature, load is higher, and relationship is closer, and development becomes
Gesture and rate are closer.The weaker information of correlation is rejected, is found out between each characteristic gas and characteristic gas and oil temperature, load
Correlated variables of the correlation more than 0.8 is as the independent variable for predicting the object gas.
Step 500:According to correlation crosstab, transformer equipment oil dissolved gas concentration prediction model is built;
The step 500 further comprises:The structure of transformer equipment oil dissolved gas concentration prediction model can profit
Tool is explored with power grid big data analysis, the algorithm of support vector machine provided using modeling tool is built in transformer oil and dissolved
Forecasting of Gas Concentration model.
Step 600:The transformer equipment oil dissolved gas concentration prediction model is optimized;
The mean square error, root-mean-square error, explanation side of the model evaluation that tool provides are explored by power grid big data analysis
The model evaluations indexs such as difference, mean absolute error and R- squared modulus, the accuracy of Integrated Evaluation Model prediction result are applicable in
Property, and the conditional parameters such as the punishment parameter of algorithm of support vector machine, kernel function type, kernel functional parameter are optimized, until
Model prediction result meets business demand.
Those skilled in the art know, on the basis of step 400 has obtained correlation crosstab, step 500 and 600
Prediction model and optimization to model, the mode that the present invention is introduced that is not limited to may be used, but existing skill may be used
Any mode in art, repeats no more.In addition, being predicted according to correlation crosstab when using any mode of the prior art
The model and technical solution optimized is each fallen within the protection domain of the application.
Step 700:Certain single features gas in the transformer equipment oil is selected to select other features as output variable
The correlative factors such as gas and the oil temperature, the load utilize the gas dissolved in oil of power trans-formers as input variable
Prediction model, predict dissolving the single features gas concentration in transformer oil.
Before using power grid big data analysis exploration tool, transformer, which can only rely on, manually examines dissolved gas dense
Degree, human cost in not only increasing, but also inevitably there is error, so the application that power grid big data analysis explores tool is necessary.
The degree of correlation between characteristic gas is determined using the method for the present invention, is reused power grid big data analysis and is explored instrument creation analysis
Model is predicted, prediction accuracy is can effectively improve.As shown in the table, first it is classified as the historical data used in prediction
Amount, second is classified as the accuracy that tool prediction is explored using power grid big data analysis, and third is classified as to be determined using the method for the present invention
After characteristic gas contingency table, the accuracy that selected parameter is predicted again, as can be seen from the table, the method for the present invention with individually
Power grid big data analysis is explored tool and is compared, and accuracy has obtained larger raising, and when historical data amount is 30 days,
Rate of accuracy reached is to a relatively accurate numerical value, and with the increase of historical data amount, and promotion effect is less apparent, due to going through
The increase of history data volume can cause the exponential type of computation complexity to increase, so considering, preferably 30 days history of the present invention
Data volume has reached an ideal balance as correlation analysis and training sample set, to whole.
Historical data amount | Power grid big data analysis explores tool | The method of the present invention |
5 | 0.78 | 0.86 |
10 | 0.79 | 0.88 |
20 | 0.81 | 0.91 |
30 | 0.83 | 0.95 |
50 | 0.83 | 0.95 |
100 | 0.84 | 0.96 |
200 | 0.85 | 0.96 |
The embodiment of the present invention reduces the workload of hand inspection transformer, avoids the error of hand inspection, Neng Gouzhun
The really variation of prediction transformer oil soluble gas in future bulk concentration, effectively identifies the Hidden fault of transformer equipment, to realize
It to the controls in advance of transformer equipment, improves work efficiency, saves manpower and materials, made for construction low-carbon society huge
Contribution.
Finally it should be noted that:The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations;Although
Present invention has been described in detail with reference to the aforementioned embodiments, it will be understood by those of ordinary skill in the art that:It still may be used
With technical scheme described in the above embodiments is modified or equivalent replacement of some of the technical features;
And these modifications or replacements, various embodiments of the present invention technical solution that it does not separate the essence of the corresponding technical solution spirit and
Range.
Claims (8)
1. a kind of method of prediction gas dissolved in oil of power trans-formers, wherein the transformer oil sets for the transformer in work
Standby oil, which is characterized in that the method includes:
Step 000:Characteristic gas is selected in the dissolved gas of the transformer oil, measures the concentration of the characteristic gas and same
The oil temperature and load of transformer described in Shi Jilu;
Step 100:Obtain the history oil temperature data and historical load data of the transformer;
Step 200:Obtain the history concentration data of the characteristic gas;
Step 300:Nondimensionalization processing is carried out to history oil temperature data and historical load data;
Step 400:Correlation and the characteristic gas between the characteristic gas and the oil temperature and the load it
Between correlation analyzed;
Step 500:The correlation determined according to step 400 builds the prediction model of the gas dissolved in oil of power trans-formers;
Step 600:The prediction model of the gas dissolved in oil of power trans-formers is optimized;
Step 700:The concentration of one of described characteristic gas is selected as output variable, select other characteristic gas concentration and
The oil temperature and the load are as input variable, using the prediction model of the gas dissolved in oil of power trans-formers, to institute
The concentration for stating one of characteristic gas is predicted.
2. the method for prediction gas dissolved in oil of power trans-formers according to claim 1, which is characterized in that
In step 000, H is selected2(hydrogen), CH4(methane), C2H6(ethane), C2H4(ethylene), C2H2(acetylene), CO (oxygen
Change carbon), CO2(carbon dioxide) is used as characteristic gas.
3. the method for prediction gas dissolved in oil of power trans-formers according to claim 1, which is characterized in that
In step 100 and step 200, the history concentration data, history oil temperature data and historical load data are nearest 15-
60 days statistical data.
4. the method for prediction gas dissolved in oil of power trans-formers according to claim 2, which is characterized in that
Step 300 further comprises:
Step 310:To H2The history concentration data of (hydrogen) carries out nondimensionalization processing:
WhereinFor density of hydrogen in i-th day transformer oil,ForAs a result, n is the transformer after nondimensionalization
The statistics number of days of density of hydrogen in oil;
Step 320:To CH4The history concentration data of (methane) carries out nondimensionalization processing:
WhereinFor methane concentration in i-th day transformer oil,ForAfter nondimensionalization as a result,For the change
The maximum value of methane concentration in depressor oil,Expression rounds up;
Step 330:To C2H6The history concentration data of (ethane) carries out nondimensionalization processing:
WhereinFor ethane concentration in i-th day transformer oil,ForAfter nondimensionalization as a result,
The maximum value and minimum value of ethane concentration in the respectively described transformer oil;
Step 340:To C2H4The history concentration data of (ethylene) carries out nondimensionalization processing:
WhereinFor ethylene concentration in i-th day transformer oil,For ethylene concentration in i-th day transformer oilAfter nondimensionalization as a result,Interval be
Step 350:To C2H2The history concentration data of (acetylene) carries out nondimensionalization processing:
WhereinFor concentration of acetylene in i-th day transformer oil,For concentration of acetylene in i-th day transformer oilResult after nondimensionalization;
Step 360:Nondimensionalization processing is carried out to the history concentration data of CO (carbon monoxide):
WhereinFor carbonomonoxide concentration in i-th day transformer oil,For carbon monoxide in i-th day transformer oil
ConcentrationAfter nondimensionalization as a result,Indicate the nature truth of a matterPower;
Step 370:To CO2The history concentration data of (carbon dioxide) carries out nondimensionalization processing:
WhereinFor gas concentration lwevel in i-th day transformer oil,For titanium dioxide in i-th day transformer oil
Concentration of carbonAs a result, m is User Defined constant after nondimensionalization, and m ∈ [1, n);Bench-scale testing the result shows that:When n is excellent
When being selected as 30 days, if m values are 17, nondimensionalization best results.
5. the method for prediction gas dissolved in oil of power trans-formers according to claim 4, which is characterized in that step 300
Further comprise:
Step 380:Nondimensionalization processing is carried out to history oil temperature data:
Wherein TEM 'iFor the oil temperature of i-th day transformer, TEM 'iFor the oil temperature TEM of i-th day transformeriNondimensionalization
Afterwards as a result, TEMmax、TEMminThe maximum value and minimum value of the respectively described transformer oil history oil temperature.
6. the method for prediction gas dissolved in oil of power trans-formers according to claim 4, which is characterized in that step 300
Further comprise:
Step 390:Nondimensionalization processing is carried out to historical load data:
Wherein ViFor the load of i-th day transformer, V 'iFor the load V of i-th day transformeriAfter nondimensionalization as a result,
VstdFor the rated load of the transformer.
7. according to a kind of method of any prediction gas dissolved in oil of power trans-formers of claim 1-6, feature
It is, the historical load is 95 probability value of historical load data.
8. according to a kind of method of any prediction gas dissolved in oil of power trans-formers of claim 1-6, feature
It is, in step 400, by correlation analysis formation characteristic gas, the correlation crosstab of oil temperature and load.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2017106299701 | 2017-07-28 | ||
CN201710629970 | 2017-07-28 |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107664690A CN107664690A (en) | 2018-02-06 |
CN107664690B true CN107664690B (en) | 2018-09-11 |
Family
ID=61097042
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710660760.9A Active CN107664690B (en) | 2017-07-28 | 2017-08-04 | A method of prediction gas dissolved in oil of power trans-formers |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107664690B (en) |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108663501A (en) * | 2017-11-29 | 2018-10-16 | 全球能源互联网研究院有限公司 | A kind of predicting model for dissolved gas in transformer oil method and system |
CN109187809A (en) * | 2018-10-27 | 2019-01-11 | 国网山东省电力公司电力科学研究院 | A kind of Gases Dissolved in Transformer Oil data generate in real time and analysis system |
CN115982557B (en) * | 2023-02-24 | 2023-05-23 | 深圳市特安电子有限公司 | Data processing system for gas concentration detection |
CN116204794B (en) * | 2023-05-04 | 2023-09-12 | 国网江西省电力有限公司电力科学研究院 | Method and system for predicting dissolved gas in transformer oil by considering multidimensional data |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103778575B (en) * | 2014-03-04 | 2017-02-15 | 国网浙江宁波市鄞州区供电公司 | Transformer state evaluation method and system |
CN104458923B (en) * | 2014-12-19 | 2016-04-13 | 深圳供电局有限公司 | Method and system for detecting warning value of content of each gas in oil in transformer |
CN104914327B (en) * | 2015-05-06 | 2018-01-30 | 北京航空航天大学 | Transformer fault maintenance Forecasting Methodology based on real-time monitoring information |
-
2017
- 2017-08-04 CN CN201710660760.9A patent/CN107664690B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN107664690A (en) | 2018-02-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107664690B (en) | A method of prediction gas dissolved in oil of power trans-formers | |
Pan et al. | Data-driven estimation of building energy consumption with multi-source heterogeneous data | |
CN107992968B (en) | Electric energy meter metering error prediction method based on integrated time series analysis technology | |
AU2018101946A4 (en) | Geographical multivariate flow data spatio-temporal autocorrelation analysis method based on cellular automaton | |
CN109685289B (en) | Method, device and system for forward prediction of blast furnace conditions | |
CN106951984B (en) | Dynamic analysis and prediction method and device for system health degree | |
CN108009667A (en) | A kind of energy demand total amount and structure prediction system | |
CN103793854A (en) | Multiple combination optimization overhead transmission line operation risk informatization assessment method | |
CN104881715B (en) | A kind of method that paper plant predicts pulp property according to waste paper proportioning | |
CN111210170A (en) | Environment-friendly management and control monitoring and evaluation method based on 90% electricity distribution characteristic index | |
CN113627735A (en) | Early warning method and system for safety risk of engineering construction project | |
Chu et al. | Online complex nonlinear industrial process operating optimality assessment using modified robust total kernel partial M-regression | |
CN107834551A (en) | A kind of power distribution network low-voltage Forecasting Methodology based on SVMs | |
CN112101761A (en) | Comprehensive evaluation method for construction effect of clean, low-carbon, safe and efficient energy system | |
CN116432123A (en) | Electric energy meter fault early warning method based on CART decision tree algorithm | |
CN104021180A (en) | Combined software defect report classification method | |
Lao et al. | The optimized multivariate grey prediction model based on dynamic background value and its application | |
CN114648235A (en) | Industrial data quality quantitative analysis report generation method based on data image | |
Song et al. | Energy consumption combination forecast of Hebei province based on the IOWA operator | |
CN111160715A (en) | BP neural network based new and old kinetic energy conversion performance evaluation method and device | |
CN103136446A (en) | Back propagation (BP) neural network landslide forecasting method based on optimized input layer | |
JP7428157B2 (en) | Analysis equipment, analysis method and program | |
CN115130924A (en) | Microgrid power equipment asset evaluation method and system under source grid storage background | |
CN105842535B (en) | A kind of main syndrome screening technique of harmonic wave based on similar features fusion | |
KR20230047731A (en) | Apparatus and method for analyzing of load prediction model based on machine learning |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |