CN111680712A - Transformer oil temperature prediction method, device and system based on similar moments in the day - Google Patents

Transformer oil temperature prediction method, device and system based on similar moments in the day Download PDF

Info

Publication number
CN111680712A
CN111680712A CN202010298787.XA CN202010298787A CN111680712A CN 111680712 A CN111680712 A CN 111680712A CN 202010298787 A CN202010298787 A CN 202010298787A CN 111680712 A CN111680712 A CN 111680712A
Authority
CN
China
Prior art keywords
similar
day
oil temperature
transformer
time
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
Application number
CN202010298787.XA
Other languages
Chinese (zh)
Other versions
CN111680712B (en
Inventor
谭风雷
陈昊
陈轩
孙小磊
佘昌佳
焦系泽
李斌
张兆君
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Maintenance Branch of State Grid Jiangsu Electric Power Co Ltd
Original Assignee
Maintenance Branch of State Grid Jiangsu Electric Power Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Maintenance Branch of State Grid Jiangsu Electric Power Co Ltd filed Critical Maintenance Branch of State Grid Jiangsu Electric Power Co Ltd
Priority to CN202010298787.XA priority Critical patent/CN111680712B/en
Publication of CN111680712A publication Critical patent/CN111680712A/en
Application granted granted Critical
Publication of CN111680712B publication Critical patent/CN111680712B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/22—Matching criteria, e.g. proximity measures
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Physics & Mathematics (AREA)
  • General Engineering & Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Evolutionary Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a transformer oil temperature prediction method, device and system based on similar moments in the day, wherein the method comprises the steps of calculating a similar day corresponding to a day to be predicted; calculating a similar time based on the calculated similar day; and finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment. The method further selects the similar time corresponding to each time of the day to be predicted in the similar day, and then predicts the oil temperature of the transformer by using the similar time, so that the prediction precision of the top layer oil temperature of the transformer can be effectively improved, a theoretical basis is provided for the insulation state evaluation of the transformer, potential hidden dangers in the transformer can be timely found, the service life of the transformer is ensured, and the power supply reliability of a power grid is improved.

Description

Transformer oil temperature prediction method, device and system based on similar moments in the day
Technical Field
The invention belongs to the technical field of transformers, and particularly relates to a transformer oil temperature prediction method, device and system based on similar moments in the day.
Background
In recent years, with the gradual construction of an extra-high voltage power grid, the scale of the power grid is continuously enlarged, and a transformer is used as a core device of a power system and has wide application in the field of power transmission and transformation of the power grid. According to the statistics of related data, about 2.6 ten thousand transformers of 220kV and above exist in the whole country. The internal insulation fault of the power transformer is caused under the condition of long-term operation under high voltage, generally cannot be judged through electric information such as voltage, current and the like, and can only be judged by utilizing physical information such as dissolved gas components in oil or oil temperature. In fact, the internal insulation fault of the transformer cannot be judged by a real-time oil temperature method in a short period, the internal insulation fault of the transformer often needs to be judged by means of the future oil temperature change trend, if the internal insulation fault can be found in advance in the initial stage of the fault, namely, the fault does not influence the running of the transformer, and relevant measures are taken in time, so that the service life of the transformer can be prolonged, the unplanned power failure time can be shortened, the stability of a power system and the reliability of social power supply are ensured, and therefore the prediction of the oil temperature of the transformer in advance becomes very.
The oil temperature of the power transformer is influenced by various factors such as weather conditions, social economy, tidal current load and the like, and has certain volatility and randomness, so that the oil temperature prediction precision is difficult to guarantee, analysis of the oil temperature variation trend is not facilitated, and the real-time monitoring of the running state of the transformer is seriously influenced. In order to ensure the safe and stable operation of the power system, the accurate prediction of the oil temperature of the transformer becomes important. In order to solve the above problems, it is urgently needed to research a transformer oil temperature prediction method with simple algorithm and high prediction accuracy.
Disclosure of Invention
Aiming at the problems, the invention provides a method, a device and a system for predicting the oil temperature of the transformer based on similar moments in the day, wherein the similar moments corresponding to the moments of the day to be predicted are further selected in the similar day, and then the oil temperature of the transformer is predicted by using the similar moments, so that the prediction precision of the top layer oil temperature of the transformer can be effectively improved, a theoretical basis is provided for the insulation state evaluation of the transformer, the potential hidden danger in the transformer can be found in time, the service life of the transformer is ensured, and the power supply reliability of a power grid is improved.
The technical purpose is achieved, the technical effect is achieved, and the invention is realized through the following technical scheme:
in a first aspect, the invention provides a method for predicting transformer oil temperature based on similar time in the day, which comprises the following steps:
calculating a similar day corresponding to the day to be predicted;
calculating a similar time based on the calculated similar day;
and finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment.
Optionally, the method for calculating the similar day includes:
determining a history sample range;
performing secondary classification on the historical samples based on meteorological factors;
and calculating a similar day corresponding to the day to be predicted based on the result of the second classification and the time factor.
Optionally, when it is determined that the history sample range is the history data of the previous N days, the performing a second classification on the history sample specifically includes the following steps:
taking the Nth day before the day to be predicted and the day to be predicted as the initial classification center CT respectivelyijkWherein i represents the classification number, j represents the iteration number, and k represents the meteorological influence factor;
calculating the distance between each historical sample and the initial classification center according to the Euclidean distance;
Figure BDA0002453202030000021
in the formula, TnkRepresents the k weather influence factor, k, of the nth day before the day to be predictedmaxRepresenting the number of meteorological influencing factors;
distributing the N historical samples to two initial classification centers according to the minimum Euclidean distance to form two new classifications and calculating the classification centers, wherein the expression of the new classification centers is as follows:
Figure BDA0002453202030000022
in the formula, NijRepresenting the number of sample days included after the jth iteration of the ith classification;
continuously iterating to form a new classification until the error square sum function value obtains the minimum value, wherein the corresponding classification is the optimal classification, the classification of the day to be predicted is called as a similar class, and the other classification is called as a dissimilar class; the expression of the sum of squared errors function is:
Figure BDA0002453202030000023
optionally, the calculating, based on the result of the second classification and the time factor, a similar day corresponding to the day to be predicted includes:
when the number of the sample days in the similar class is more than or equal to X days, selecting the X sample days closest to the day to be predicted as the similar days;
and when the number of the sample days in the similar class is less than X days, all the sample days in the similar class are taken as the similar days.
Optionally, the method for calculating the similar time includes:
determining meteorological factors related to the oil temperature of the transformer;
calculating the correlation between meteorological factors and the transformer oil temperature at each moment;
and calculating the similar time based on the calculated correlation and the set similar time judgment condition.
Optionally, the calculation formula of the correlation between the meteorological factors and the transformer oil temperature at each time is as follows:
Rnm＝RRWnm×RRSnm×RRFnm×RRQnm(4)
in the formula, RRWnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the temperature factor at the mth time; RRSnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the humidity factor at the mth time; RRFnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the wind speed factor at the mth time; RRQnmAnd the intermediate variable represents the correlation degree of the oil temperature of the transformer at the nth time and the air pressure factor at the mth time.
Optionally, the similar time determination condition is:
Rnm≥Rnn(5)
in the formula, RnmRepresenting the correlation degree, R, of the oil temperature of the transformer at the nth time and meteorological factors at the mth timennRepresenting the correlation degree of the transformer oil temperature at the nth time and meteorological factors at the nth time;
and when the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is greater than or equal to the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the nth time, the mth time is considered as the similar time of the nth time.
Optionally, when the similar time at the nth time of the day to be predicted is the mth time, the number of the similar times is BnThe oil temperature prediction method of the transformer comprises the following steps:
selecting meteorological factors of which the correlation degree with the oil temperature of the transformer at the nth time is greater than or equal to a set threshold value as main meteorological factor variables;
taking the main meteorological factor variable corresponding to the mth moment of each similar day as an input set of a neural network, taking the transformer oil temperature at the nth moment of each similar day as an output set, and training the neural network after reasonably setting the number of neuron layers, the number of single-layer neurons, a single-layer neuron transmission function, training parameters and a training function of the neural network;
inputting the main meteorological factor variable corresponding to the mth moment of the day to be predicted into the trained neural network, wherein the output value of the neural network at the moment is the predicted value P of the transformer oil temperature at the nth moment of the day to be predicted based on the similar moment mnm；
Number of similar moments B when n isnWhen the value is 1, the predicted value P of the transformer oil temperature at the nth time of the day to be predictedn＝Pnm(ii) a Number of similar moments B when n isnWhen the temperature is more than 1, the predicted value of the transformer oil temperature at the nth time of the day to be predicted is a predicted value P based on each similar time mnmThe expression of linear weighted value of (a) is as follows:
Figure BDA0002453202030000031
wherein R isnmAnd the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is shown.
In a second aspect, the present invention provides a device for predicting transformer oil temperature based on similar time in the day, including:
the first calculating unit is used for calculating a similar day corresponding to the day to be predicted;
the second calculation unit is used for screening out similar moments based on the calculated similar days;
and the prediction unit is used for finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment.
In a third aspect, the present invention provides a system for predicting transformer oil temperature based on similar time in the day, including: a storage medium and a processor;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of the first aspects.
The invention has the beneficial effects that:
on the basis of fully considering meteorological and time factors, the method further selects the similar time in the similar day, and then predicts the oil temperature of the transformer by using the similar time, improves the prediction precision, and provides a theoretical basis for the insulation state evaluation of the transformer and the maximum utilization of the capacity of the transformer. Meanwhile, whether the oil temperature of the transformer exceeds the limit temperature or not can be judged in time, a basis is provided for the development of related work of dispatching operation, and the power supply reliability of the system is ensured.
Furthermore, on the basis of fully considering meteorological and time factors, the method selects the similar days based on two categories, and ensures the effectiveness of similar day selection.
Furthermore, the method selects the similar time based on the correlation between the meteorological factors and the transformer oil temperature, and ensures the effectiveness of similar time selection.
Drawings
FIG. 1 is an overall flow chart of an algorithm of a transformer oil temperature prediction method of the present invention;
FIG. 2 is a flow chart of similar day selection for the transformer oil temperature prediction method of the present invention;
FIG. 3 is a flow chart of similar time selection for the transformer oil temperature prediction method of the present invention;
FIG. 4 is a flow chart of oil temperature prediction for a transformer oil temperature prediction method of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The following detailed description of the principles of the invention is provided in connection with the accompanying drawings.
On the basis of fully considering meteorological and time factors, the method further selects the similar time in the similar day, and then predicts the oil temperature of the transformer by using the similar time, improves the prediction precision, and provides a theoretical basis for the insulation state evaluation of the transformer and the maximum utilization of the capacity of the transformer. Meanwhile, whether the oil temperature of the transformer exceeds the limit temperature or not can be judged in time, a basis is provided for the development of related work of dispatching operation, and the power supply reliability of the system is ensured.
Example 1
As shown in fig. 1, an overall flowchart of a transformer oil temperature prediction method based on similar time in the day in the embodiment of the present invention is shown. The transformer oil temperature prediction method based on similar moments in the day comprises the following steps:
(1) calculating a similar day corresponding to the day to be predicted;
in a specific implementation manner of the embodiment of the present invention, as shown in fig. 2, the step (1) specifically includes the following steps:
step 1.1: a history sample range is determined.
Considering that when the selection range of the historical samples is large, the condition that meteorological factors are similar but time correlation is low exists, and the effectiveness of similar day selection is influenced; when the selection range of the historical samples is small, the historical samples with high correlation degree of meteorological factors are difficult to select and obtain. In a specific application process, 30 days before the day to be predicted can be selected as a history sample in combination with the time variation trend.
Step 1.2: the historical samples are classified two times based on meteorological factors.
And performing secondary classification on the historical samples based on meteorological factors, wherein 10 types of meteorological factors such as daily maximum temperature, daily average temperature, daily minimum temperature, daily maximum humidity, daily average humidity, daily minimum humidity, daily average wind speed, daily average air pressure, daily rainfall, sunlight intensity and the like are selected as classification factors, and the historical samples are divided into two types, wherein the type of the day to be predicted is called a similar type, and the other type is called a non-similar type.
The following is described in detail with reference to a specific implementation, as shown in fig. 2:
(1) selecting a historical sample range according to the day to be predicted, and carrying out normalization processing on 10 types of meteorological factors such as the maximum daily temperature, the average daily temperature and the like, wherein the classification number is set to be 2, and the iteration number is set to be R.
(2) Selecting the 30 th day before the day to be predicted and the day to be predicted as an initial classification center CTijkIn the formula, i represents the classification number,j represents the number of iterations and k represents the weather influencing factor.
(3) And calculating the distance between each historical sample and the initial classification center according to the Euclidean distance:
Figure BDA0002453202030000051
in the formula, TnkRepresenting the k-th influencing factor of the nth day before the day to be predicted.
(4) Distributing 30 historical samples to two initial classification centers according to the minimum Euclidean distance to form two new classifications and calculating the classification centers, wherein the expression of the new classification centers is as follows:
Figure BDA0002453202030000052
in the formula, NijRepresenting the number of sample days included after the jth iteration of the ith classification.
(5) Continuously iterating to form a new classification until the error square sum function value obtains the minimum value, wherein the corresponding classification is the optimal classification, the classification of the day to be predicted is called as a similar class, and the other classification is called as a dissimilar class; the expression of the sum of squared errors function is:
Figure BDA0002453202030000061
step 1.3: and calculating a similar day corresponding to the day to be predicted based on the result of the second classification and the time factor.
Considering that only meteorological factors are considered in the second classification, considering time factors below, selecting similar days from similar classes, and selecting X sample days closest to the day to be predicted as the similar days when the number of the sample days in the similar classes is more than or equal to X (X can be 10); and when the number of the sample days in the similar class is less than X days, all the sample days in the similar class are taken as the similar days.
(2) Calculating a similar time based on the calculated similar day;
in a specific implementation manner of the embodiment of the present invention, as shown in fig. 3, the step (2) specifically includes the following steps:
step 2.1: determining meteorological factors related to the oil temperature of the transformer;
according to the principle of correlation, when the degree of correlation is greater than or equal to 0.4, the correlation is more than or equal to a medium level; when the correlation degree is less than 0.4, weak correlation is shown, so that only meteorological factors with the correlation degree of the transformer oil temperature being more than or equal to 0.4 are studied when similar moments are analyzed. Definition RWnmRepresenting the correlation degree of the oil temperature of the transformer at the nth time and the temperature factor at the mth time, RRWnmRepresenting the correlation intermediate variable between the oil temperature of the transformer at the nth time and the temperature factor at the mth time when RW isnmWhen the ratio is more than or equal to 0.4, RRWnm＝RWnmWhen RW isnm<At 0.4, RRWnm＝1；RSnmRepresenting the correlation degree of the oil temperature of the transformer at the nth time and the humidity factor at the mth time, RRSnmRepresenting the correlation intermediate variable of the oil temperature of the transformer at the nth time and the humidity factor at the mth time when RSnmWhen the ratio is more than or equal to 0.4, RRSnm＝RSnmWhen RSnm<At 0.4, RRSnm＝1；RFnmRepresenting the correlation degree, RRF, of the oil temperature of the transformer at the nth time and the wind speed factor at the mth timenmRepresenting the correlation intermediate variable of the oil temperature of the transformer at the nth time and the wind speed factor at the mth time when the RF is appliednmWhen the ratio is more than or equal to 0.4, RRFnm＝RFnmWhen RFnm<At 0.4, RRFnm＝1；RQnmRepresenting the correlation degree of the oil temperature of the transformer at the nth time and the air pressure factor at the mth time, RRQnmRepresenting the correlation intermediate variable of the oil temperature of the transformer at the nth time and the air pressure factor at the mth time, when the correlation intermediate variable is RQnmWhen the ratio is more than or equal to 0.4, RRQnm＝RQnmWhen it is RQnm<At 0.4, RRQnm＝1。
Step 2.2: and calculating the correlation between the meteorological factors and the transformer oil temperature at each moment.
According to the selected meteorological factors, the correlation degree of the meteorological factors and the transformer oil temperature at a certain moment can be obtained:
Rnm＝RRWnm×RRSnm×RRFnm×RRQnm(4)
in the formula, RnmAnd the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is shown.
Step 2.3: and calculating the similar time based on the calculated correlation and the set similar time judgment condition.
Defining that when the correlation degree of the transformer oil temperature at the nth moment and the meteorological factors at the mth moment is greater than or equal to the correlation degree of the transformer oil temperature at the nth moment and the meteorological factors at the nth moment, the mth moment is considered as a similar moment of the nth moment, and the specific judgment conditions are as follows:
Rnm≥Rnn(5)
wherein R isnmRepresenting the correlation degree, R, of the oil temperature of the transformer at the nth time and meteorological factors at the mth timennAnd the correlation degree of the transformer oil temperature at the nth time and meteorological factors at the nth time is shown.
(3) Based on the oil temperature data of the transformer at the similar time, the oil temperature prediction of the transformer is completed;
in a specific implementation manner of the embodiment of the present invention, the step (3) specifically includes:
because a plurality of similar days generally exist on a day to be predicted, and each time of the day to be predicted also corresponds to a plurality of similar times, the BP neural network is adopted to predict the oil temperature of the transformer, the number of the similar days of the day to be predicted is firstly determined as A, the number of the similar times of the nth time is determined as the mth time, and the number of the similar times is BnThen, as shown in fig. 4, the specific implementation process is as follows:
(1) and selecting factors with the correlation degree of the transformer oil temperature at the nth time being more than or equal to 0.4 from meteorological factors such as temperature, humidity, wind speed and air pressure as main meteorological factor variables.
(2) And (3) taking the main meteorological factor variable corresponding to the mth moment of each similar day as an input set of the BP neural network, taking the transformer oil temperature at the nth moment of each similar day as an output set, reasonably setting the number of neuron layers, the number of single-layer neurons, the single-layer neuron transmission function, the training parameters and the training function of the BP neural network, and then training the BP neural network.
(3) Day to be predictedInputting the main meteorological factor variable corresponding to the mth moment into the trained BP neural network, wherein the output value of the neural network at the moment is the predicted value P of the transformer oil temperature at the nth moment of the day to be predicted based on the similar moment mnm。
(4) Number of similar moments B when n isnWhen the value is 1, the predicted value P of the transformer oil temperature at the nth time of the day to be predictedn＝Pnm(ii) a Number of similar moments B when n isn>1, the predicted value of the transformer oil temperature at the nth time of the day to be predicted is a predicted value P based on each similar time mnmThe expression of linear weighted value of (a) is as follows:
Figure BDA0002453202030000071
wherein R isnmAnd the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is shown.
Example 2
Based on the same inventive concept as embodiment 1, an embodiment of the present invention provides a transformer oil temperature prediction device based on similar time in the day, including:
the first calculating unit is used for calculating a similar day corresponding to the day to be predicted;
the second calculation unit is used for screening out similar moments based on the calculated similar days;
and the prediction unit is used for finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment.
In a specific implementation manner of the embodiment of the present invention, when determining that the range of the historical samples is the historical data of the previous N days, performing a second classification on the historical samples specifically includes the following steps:
taking the Nth day before the day to be predicted and the day to be predicted as the initial classification center CT respectivelyijkWherein i represents the classification number, j represents the iteration number, and k represents the meteorological influence factor;
calculating the distance between each historical sample and the initial classification center according to the Euclidean distance;
Figure BDA0002453202030000081
in the formula, TnkRepresents the k weather influence factor, k, of the nth day before the day to be predictedmaxRepresenting the number of meteorological influencing factors;
distributing the N historical samples to two initial classification centers according to the minimum Euclidean distance to form two new classifications and calculating the classification centers, wherein the expression of the new classification centers is as follows:
Figure BDA0002453202030000082
in the formula, NijRepresenting the number of sample days included after the jth iteration of the ith classification;
continuously iterating to form a new classification until the error square sum function value obtains the minimum value, wherein the corresponding classification is the optimal classification, the classification of the day to be predicted is called as a similar class, and the other classification is called as a dissimilar class; the expression of the sum of squared errors function is:
Figure BDA0002453202030000083
in a specific implementation manner of the embodiment of the present invention, the calculating a similar day corresponding to a day to be predicted based on the result of the second classification and the time factor includes:
when the number of the sample days in the similar class is more than or equal to X days, selecting the X sample days closest to the day to be predicted as the similar days;
and when the number of the sample days in the similar class is less than X days, all the sample days in the similar class are taken as the similar days.
In a specific implementation manner of the embodiment of the present invention, the method for calculating the similar time includes:
determining meteorological factors related to the oil temperature of the transformer;
calculating the correlation between meteorological factors and the transformer oil temperature at each moment;
and calculating the similar time based on the calculated correlation and the set similar time judgment condition.
In a specific implementation manner of the embodiment of the present invention, a calculation formula of the correlation between the meteorological factors and the transformer oil temperature at each time is as follows:
Rnm＝RRWnm×RRSnm×RRFnm×RRQnm(4)
in the formula, RRWnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the temperature factor at the mth time; RRSnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the humidity factor at the mth time; RRFnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the wind speed factor at the mth time; RRQnmAnd the intermediate variable represents the correlation degree of the oil temperature of the transformer at the nth time and the air pressure factor at the mth time.
In a specific implementation manner of the embodiment of the present invention, the similar time determination condition is:
Rnm≥Rnn(5)
in the formula, RnmRepresenting the correlation degree, R, of the oil temperature of the transformer at the nth time and meteorological factors at the mth timennRepresenting the correlation degree of the transformer oil temperature at the nth time and meteorological factors at the nth time;
and when the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is greater than or equal to the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the nth time, the mth time is considered as the similar time of the nth time.
In a specific implementation manner of the embodiment of the present invention, the nth time of the day to be predicted is the mth time, and the number of the similar times is BnThe oil temperature prediction method of the transformer comprises the following steps:
selecting meteorological factors of which the correlation degree with the oil temperature of the transformer at the nth time is greater than or equal to a set threshold value as main meteorological factor variables;
taking the main meteorological factor variable corresponding to the mth moment of each similar day as an input set of a neural network, taking the transformer oil temperature at the nth moment of each similar day as an output set, and training the neural network after reasonably setting the number of neuron layers, the number of single-layer neurons, a single-layer neuron transmission function, training parameters and a training function of the neural network;
inputting the main meteorological factor variable corresponding to the mth moment of the day to be predicted into the trained neural network, wherein the output value of the neural network at the moment is the predicted value P of the transformer oil temperature at the nth moment of the day to be predicted based on the similar moment mnm；
Number of similar moments B when n isnWhen the value is 1, the predicted value P of the transformer oil temperature at the nth time of the day to be predictedn＝Pnm(ii) a Number of similar moments B when n isn>1, the predicted value of the transformer oil temperature at the nth time of the day to be predicted is a predicted value P based on each similar time mnmThe expression of linear weighted value of (a) is as follows:
Figure BDA0002453202030000091
the correlation between the transformer oil temperature at the nth time and meteorological factors at the mth time is shown.
Example 3
Based on the same inventive concept as embodiment 1, an embodiment of the present invention provides a transformer oil temperature prediction system based on similar time in the day, including: a storage medium and a processor;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method of any of embodiment 1.
As will be appreciated by one skilled in the art, embodiments of the present application may be provided as a method, system, or computer program product. Accordingly, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present application may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The present application is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the application. It will be understood that each flow and/or block of the flow diagrams and/or block diagrams, and combinations of flows and/or blocks in the flow diagrams and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the present invention has been described with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, which are illustrative and not restrictive, and it will be apparent to those skilled in the art that various changes and modifications can be made therein without departing from the spirit and scope of the invention as defined in the appended claims.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. It will be understood by those skilled in the art that the present invention is not limited to the embodiments described above, which are described in the specification and illustrated only to illustrate the principle of the present invention, but that various changes and modifications may be made therein without departing from the spirit and scope of the present invention, which fall within the scope of the invention as claimed. The scope of the invention is defined by the appended claims and equivalents thereof.

Claims (10)

1. A transformer oil temperature prediction method based on similar moments in the day is characterized by comprising the following steps:
calculating a similar day corresponding to the day to be predicted;
calculating a similar time based on the calculated similar day;
and finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment.
2. The method for predicting the oil temperature of the transformer based on the similar time in the day according to claim 1, wherein the calculating method of the similar day comprises:
determining a history sample range;
performing secondary classification on the historical samples based on meteorological factors;
and calculating a similar day corresponding to the day to be predicted based on the result of the second classification and the time factor.
3. The method for predicting the oil temperature of the transformer based on the similar time in the day as claimed in claim 2, wherein when the historical sample range is determined to be the historical data of the previous N days, the secondary classification is performed on the historical sample, and specifically comprises the following steps:
taking the Nth day before the day to be predicted and the day to be predicted as the initial classification center CT respectivelyijkWherein i represents the classification number, j represents the iteration number, and k represents the meteorological influence factor;
calculating the distance between each historical sample and the initial classification center according to the Euclidean distance;
Figure FDA0002453202020000011
in the formula, TnkRepresents the k weather influence factor, k, of the nth day before the day to be predictedmaxRepresenting the number of meteorological influencing factors;
distributing the N historical samples to two initial classification centers according to the minimum Euclidean distance to form two new classifications and calculating the classification centers, wherein the expression of the new classification centers is as follows:
Figure FDA0002453202020000012
in the formula, NijRepresenting the number of sample days included after the jth iteration of the ith classification;
continuously iterating to form a new classification until the error square sum function value obtains the minimum value, wherein the corresponding classification is the optimal classification, the classification of the day to be predicted is called as a similar class, and the other classification is called as a dissimilar class; the expression of the sum of squared errors function is:
Figure FDA0002453202020000013
4. the method for predicting the oil temperature of the transformer based on the similar time in the day according to claim 3,
the calculating of the similar day corresponding to the day to be predicted based on the result of the two classifications and the time factor comprises the following steps:
when the number of the sample days in the similar class is more than or equal to X days, selecting the X sample days closest to the day to be predicted as the similar days;
and when the number of the sample days in the similar class is less than X days, all the sample days in the similar class are taken as the similar days.
5. The method for predicting the transformer oil temperature based on the similar time in the day according to claim 1, wherein the method for calculating the similar time comprises the following steps:
determining meteorological factors related to the oil temperature of the transformer;
calculating the correlation between meteorological factors and the transformer oil temperature at each moment;
and calculating the similar time based on the calculated correlation and the set similar time judgment condition.
6. The method for predicting the oil temperature of the transformer based on the similar time in the day as claimed in claim 5, wherein the calculation formula of the correlation degree of the meteorological factors and the oil temperature of the transformer at each time is as follows:
Rnm＝RRWnm×RRSnm×RRFnm×RRQnm(4)
in the formula, RRWnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the temperature factor at the mth time; RRSnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the humidity factor at the mth time; RRFnmRepresenting a correlation degree intermediate variable of the oil temperature of the transformer at the nth time and the wind speed factor at the mth time; RRQnmAnd the intermediate variable represents the correlation degree of the oil temperature of the transformer at the nth time and the air pressure factor at the mth time.
7. The method for predicting the oil temperature of the transformer based on the similar time in the day according to claim 5, wherein the similar time determination conditions are as follows:
Rnm≥Rnn(5)
in the formula, RnmRepresenting the correlation degree, R, of the oil temperature of the transformer at the nth time and meteorological factors at the mth timennRepresenting the correlation degree of the transformer oil temperature at the nth time and meteorological factors at the nth time;
and when the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is greater than or equal to the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the nth time, the mth time is considered as the similar time of the nth time.
8. The method for predicting the oil temperature of the transformer based on the similar moments in the day as claimed in claim 1, wherein when the similar moment at the nth moment of the day to be predicted is the mth moment, the number of the similar moments is BnThe oil temperature prediction method of the transformer comprises the following steps:
selecting meteorological factors of which the correlation degree with the oil temperature of the transformer at the nth time is greater than or equal to a set threshold value as main meteorological factor variables;
taking the main meteorological factor variable corresponding to the mth moment of each similar day as an input set of a neural network, taking the transformer oil temperature at the nth moment of each similar day as an output set, and training the neural network after reasonably setting the number of neuron layers, the number of single-layer neurons, a single-layer neuron transmission function, training parameters and a training function of the neural network;
inputting the main meteorological factor variable corresponding to the mth moment of the day to be predicted into the trained neural network, wherein the output value of the neural network at the moment is the predicted value P of the transformer oil temperature at the nth moment of the day to be predicted based on the similar moment mnm；
Number of similar moments B when n isnWhen the value is 1, the predicted value P of the transformer oil temperature at the nth time of the day to be predictedn＝Pnm；
Number of similar moments B when n isnWhen the temperature is more than 1, the predicted value of the transformer oil temperature at the nth time of the day to be predicted is a predicted value P based on each similar time mnmThe expression of linear weighted value of (a) is as follows:
Figure FDA0002453202020000031
wherein R isnmAnd the correlation degree of the transformer oil temperature at the nth time and the meteorological factors at the mth time is shown.
9. A transformer oil temperature prediction device based on similar moments in the day is characterized by comprising the following components:
the first calculating unit is used for calculating a similar day corresponding to the day to be predicted;
the second calculation unit is used for screening out similar moments based on the calculated similar days;
and the prediction unit is used for finishing the oil temperature prediction of the transformer based on the oil temperature data of the transformer at the similar moment.
10. A transformer oil temperature prediction system based on similar moments in the day is characterized by comprising: a storage medium and a processor;
the storage medium is used for storing instructions;
the processor is configured to operate in accordance with the instructions to perform the steps of the method according to any one of claims 1 to 8.
CN202010298787.XA 2020-04-16 2020-04-16 Method, device and system for predicting oil temperature of transformer based on similar time in day Active CN111680712B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010298787.XA CN111680712B (en) 2020-04-16 2020-04-16 Method, device and system for predicting oil temperature of transformer based on similar time in day

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010298787.XA CN111680712B (en) 2020-04-16 2020-04-16 Method, device and system for predicting oil temperature of transformer based on similar time in day

Publications (2)

Publication Number Publication Date
CN111680712A true CN111680712A (en) 2020-09-18
CN111680712B CN111680712B (en) 2023-06-06

Family

ID=72433318

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010298787.XA Active CN111680712B (en) 2020-04-16 2020-04-16 Method, device and system for predicting oil temperature of transformer based on similar time in day

Country Status (1)

Country Link
CN (1) CN111680712B (en)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112347695A (en) * 2020-10-29 2021-02-09 国网江苏省电力有限公司检修分公司 Method and system for predicting top-layer oil temperature of transformer
CN112529306A (en) * 2020-12-16 2021-03-19 国网江苏省电力有限公司检修分公司 Latent defect diagnosis method and device for phase modulator based on oil temperature estimation of oil system
CN112541634A (en) * 2020-12-16 2021-03-23 国网江苏省电力有限公司检修分公司 Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104299057A (en) * 2014-10-17 2015-01-21 国家电网公司 Method for predicting top layer oil temperature of oil-immersed transformer based on multiple factors
WO2015172560A1 (en) * 2014-05-16 2015-11-19 华南理工大学 Central air conditioner cooling load prediction method based on bp neural network
CN108564192A (en) * 2017-12-29 2018-09-21 河海大学 A kind of short-term photovoltaic power prediction technique based on meteorological factor weight similar day
CN110148068A (en) * 2019-05-23 2019-08-20 福州大学 One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method
CN110852492A (en) * 2019-10-25 2020-02-28 东北电力大学 Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015172560A1 (en) * 2014-05-16 2015-11-19 华南理工大学 Central air conditioner cooling load prediction method based on bp neural network
CN104299057A (en) * 2014-10-17 2015-01-21 国家电网公司 Method for predicting top layer oil temperature of oil-immersed transformer based on multiple factors
CN108564192A (en) * 2017-12-29 2018-09-21 河海大学 A kind of short-term photovoltaic power prediction technique based on meteorological factor weight similar day
CN110148068A (en) * 2019-05-23 2019-08-20 福州大学 One kind being based on meteorological data similarity analysis and LSTM neural fusion photovoltaic plant ultra-short term power forecasting method
CN110852492A (en) * 2019-10-25 2020-02-28 东北电力大学 Photovoltaic power ultra-short-term prediction method for finding similarity based on Mahalanobis distance

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
L. RAEISIAN ET AL.: "Prediction of Oil Hotspot Temperature in a Distribution Transformer by CFD Method" *
卢舟鑫 等: "基于多时段综合相似日的光伏发电功率预测", 《电源技术》 *
宋人杰 等: "基于相似日和WNN的光伏发电功率超短期预测模型", 《电测与仪表》 *
程泽 等: "基于相似时刻的光伏出力概率分布估计方法", 《电网技术》 *
陈锦铭 等: "变压器顶层油温短期预测模型研究及应用", 《电气应用》 *

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112347695A (en) * 2020-10-29 2021-02-09 国网江苏省电力有限公司检修分公司 Method and system for predicting top-layer oil temperature of transformer
CN112347695B (en) * 2020-10-29 2023-12-01 国网江苏省电力有限公司检修分公司 Method and system for predicting top layer oil temperature of transformer
CN112529306A (en) * 2020-12-16 2021-03-19 国网江苏省电力有限公司检修分公司 Latent defect diagnosis method and device for phase modulator based on oil temperature estimation of oil system
CN112541634A (en) * 2020-12-16 2021-03-23 国网江苏省电力有限公司检修分公司 Top layer oil temperature prediction and false fire alarm discrimination method, device and storage medium
CN112529306B (en) * 2020-12-16 2023-12-01 国网江苏省电力有限公司检修分公司 Camera latent defect diagnosis method and device based on oil temperature estimation of oil system
CN112541634B (en) * 2020-12-16 2024-03-15 国网江苏省电力有限公司检修分公司 Method and device for predicting top-layer oil temperature and discriminating false alarm and storage medium

Also Published As

Publication number Publication date
CN111680712B (en) 2023-06-06

Similar Documents

Publication Publication Date Title
CN111680712B (en) Method, device and system for predicting oil temperature of transformer based on similar time in day
CN112215442B (en) Method, system, device and medium for predicting short-term load of power system
CN108241901B (en) Transformer early warning evaluation method and device based on prediction data
CN111525587B (en) Reactive load situation-based power grid reactive voltage control method and system
CN106529732A (en) Carbon emission efficiency prediction method based on neural network and random frontier analysis
CN105468850A (en) Multi-residual error regression prediction algorithm based electronic product degradation trend prediction method
CN113449919B (en) Power consumption prediction method and system based on feature and trend perception
CN112365056A (en) Electrical load joint prediction method and device, terminal and storage medium
CN110837915A (en) Low-voltage load point prediction and probability prediction method for power system based on hybrid integrated deep learning
CN114066071A (en) Power parameter optimization method based on energy consumption, terminal equipment and storage medium
CN115358437A (en) Power supply load prediction method based on convolutional neural network
CN112529053A (en) Short-term prediction method and system for time sequence data in server
CN111815039A (en) Weekly scale wind power probability prediction method and system based on weather classification
CN114154716B (en) Enterprise energy consumption prediction method and device based on graph neural network
CN116050595A (en) Attention mechanism and decomposition mechanism coupled runoff amount prediction method
CN104866679A (en) Mechanical characteristic prediction method of switching device
Gao et al. How to improve the application potential of deep learning model in HVAC fault diagnosis: Based on pruning and interpretable deep learning method
CN109523077B (en) Wind power prediction method
CN113742993A (en) Method, device, equipment and storage medium for predicting life loss of dry-type transformer
CN113033898A (en) Electrical load prediction method and system based on K-means clustering and BI-LSTM neural network
CN117113086A (en) Energy storage unit load prediction method, system, electronic equipment and medium
CN116151799A (en) BP neural network-based distribution line multi-working-condition fault rate rapid assessment method
CN114358382A (en) Power system equivalent inertia probability prediction method, device, product and storage medium
CN114372558A (en) Residential electricity load prediction method, medium and equipment based on multi-model fusion
Khan et al. Short term load and price forecasting using tuned parameters for K-nearest neighbors

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