CN116720622A - Method for predicting metering error value of current transformer - Google Patents
Method for predicting metering error value of current transformer Download PDFInfo
- Publication number
- CN116720622A CN116720622A CN202310699195.2A CN202310699195A CN116720622A CN 116720622 A CN116720622 A CN 116720622A CN 202310699195 A CN202310699195 A CN 202310699195A CN 116720622 A CN116720622 A CN 116720622A
- Authority
- CN
- China
- Prior art keywords
- error
- temperature
- data
- component
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012549 training Methods 0.000 claims abstract description 14
- 238000011156 evaluation Methods 0.000 claims abstract description 13
- YHXISWVBGDMDLQ-UHFFFAOYSA-N moclobemide Chemical compound C1=CC(Cl)=CC=C1C(=O)NCCN1CCOCC1 YHXISWVBGDMDLQ-UHFFFAOYSA-N 0.000 claims abstract description 7
- 238000003062 neural network model Methods 0.000 claims abstract description 4
- 230000000737 periodic effect Effects 0.000 claims description 27
- 238000013528 artificial neural network Methods 0.000 claims description 6
- 230000007787 long-term memory Effects 0.000 claims description 3
- 238000009499 grossing Methods 0.000 description 4
- 238000005259 measurement Methods 0.000 description 4
- 230000007613 environmental effect Effects 0.000 description 3
- 230000015556 catabolic process Effects 0.000 description 2
- 238000000354 decomposition reaction Methods 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical group [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 1
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005674 electromagnetic induction Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 230000001932 seasonal effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000004804 winding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/02—Testing or calibrating of apparatus covered by the other groups of this subclass of auxiliary devices, e.g. of instrument transformers according to prescribed transformation ratio, phase angle, or wattage rating
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/24—Querying
- G06F16/245—Query processing
- G06F16/2458—Special types of queries, e.g. statistical queries, fuzzy queries or distributed queries
- G06F16/2474—Sequence data queries, e.g. querying versioned data
-
- 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
- G06N3/044—Recurrent networks, e.g. Hopfield networks
- G06N3/0442—Recurrent networks, e.g. Hopfield networks characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU]
-
- 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
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
Abstract
The invention provides a method for predicting a metering error value of a current transformer, which comprises the following steps: taking an error evaluation value obtained by a mutual inductor error evaluation algorithm as historical training data, and decomposing a time sequence by using an STL method to obtain three components; collecting various influence factor data, calculating the influence of temperature on a trend component and a period component, and obtaining the trend component and the period component after the influence of the stripping temperature for subsequent prediction; predicting a trend component and a period component of the future time interval to be measured after stripping the temperature influence factors by adopting an arima model algorithm, wherein the result is added with a predicted temperature additional error of the time to be measured to be used as a trend component and period component prediction result; respectively predicting other influencing factors, and importing a trained LSTM neural network model to predict future time period residual components; and combining the separately predicted error components to generate a future transformer error prediction result.
Description
Technical Field
The invention relates to the technical field of online monitoring of electric power metering, in particular to a metering error value prediction method of a current transformer.
Background
As an important component of the electric energy metering device, the accuracy and reliability of the metering performance of the mutual inductor are directly related to fairness and fairness of electric energy trade settlement. The Current Transformer (CT) is an instrument for converting primary side large current into secondary side small current according to the electromagnetic induction principle to measure, and consists of a closed iron core and a winding.
In the actual running process of the CT, the transformer error is influenced by the acquisition principle, the severe environment and the like, and measurement deviation out-of-limit can occur in the service life of the transformer, so that accurate and rapid diagnosis can be carried out when the measurement error is out of tolerance, further, timely prediction is required to be carried out on the degradation trend of the CT measurement error, so that relevant operation maintainers can arrange maintenance work, and if the state degradation of the transformer cannot be found timely, the operation of a power grid can be influenced.
Zhang Zhu on the basis of CVT metering error state evaluation, Q statistics representing the CVT metering error state and a statistical control threshold thereof are selected as a prediction object in the capacitive voltage transformer metering error state evaluation and prediction method research, and the state prediction of the CVT metering error is mapped into the state prediction of the physical correlation of the power grid information, so that the influence of power grid fluctuation on an evaluation result is eliminated. Establishing a CVT metering error state prediction model based on Q-ARMA according to a time sequence analysis method: the method for establishing the prediction model is researched, and comprises the steps of stability processing of a time sequence, identification of model types, pricing principles of the model, parameter estimation of the model and the like. The result shows that the CVT metering error state prediction model based on the Q-ARMA can predict the change trend of the metering error state of the CVT at the next sampling moment or in a future time period.
In order to avoid inaccuracy of information sources of a secondary information system, reduce loss of electric energy metering and ensure normal operation of a measurement and control protection device, how to predict future variation trend of CT errors so as to early warn risks of CT occurrence in advance is a technical problem.
Disclosure of Invention
The invention aims to provide a method for predicting the metering error value of a current transformer and accurately predicting the future error of the current transformer aiming at the defects of the prior art.
In order to achieve the above purpose, the invention adopts the following technical scheme:
the invention provides a method for predicting a metering error value of a current transformer, which comprises the following steps:
s1, taking an error evaluation value obtained by a mutual inductor error evaluation algorithm as historical training data, and decomposing a time sequence by using an STL method to obtain three components;
s2, collecting data of various influencing factors, calculating influences of temperature on trend components and periodic components, and obtaining trend components and periodic components after the influence of the stripping temperature for subsequent prediction;
s3, predicting trend components and periodic components of the future time interval to be measured after stripping temperature influence factors by adopting an arima model algorithm, adding an additional error corresponding to the predicted temperature of the time interval to be measured as a trend component and periodic component prediction result, and introducing the predicted influence factors into an LSTM neural network model to predict future time interval residual error components;
s4, combining the error components of the separate prediction to generate a future transformer error prediction result.
Further, the S1 specifically includes: taking an error estimation value obtained by a mutual inductor error estimation algorithm as historical training data, wherein the acquisition frequency is 1 per hour, the data quantity is N, a predicted data sequence is formed, and the predicted data sequence is recorded as: y is Y V ;
The STL method is adopted to decompose the time corresponding to the collected historical transformer error sequence data into three components, namely a trend component, a period component and a residual error component, namely:
Y v =T v +S v +R v (1)
wherein T is v 、S v 、R v The trend component, the periodic component and the residual component of the v moment are respectively.
Further, the specific steps of S2 are as follows:
s201, collecting time data corresponding to each influence factor, wherein the time data is once per hour;
s202, recording a temperature time sequence of acquired contemporaneous historical training data as TEM, a humidity time sequence as HUM, a primary frequency time sequence as FRE, a load time sequence as LOA, a VA, a vibration amplitude time sequence as VIB, a mm and a magnetic field time sequence as MAG;
s203, fitting the TEM by using a least square method to obtain a fitting curve y TEM (v)=a 0 +a 1 x+...+a n x n The fitted n-th-order item curve; taking n=3 as an influence factor of temperature on T;
s204, recording the calculated average value of the data points of the historical time series data TEM at 6 days asThe average value of the data points at 12 daily in TEM was calculated and recorded as +.>Taking 24h as a period, and obtaining a fitting curve S of the temperature under the influence of the period according to the lowest temperature at 6 am and the highest temperature at 12 am TEM (v):
Wherein V' is the time of the day of V;
s205, calculating trend component T and periodic component S respectively, stripping temperature factor and then predicting trend component T p Periodic component S p The temperature additional error formula is:
wherein B is rated load;is a preset constant; a, a c Is the temperature coefficient; Δt is the difference between the temperature of the measuring point and 20 ℃; omega n Is the rated angular frequency; i 1 Is the primary current; the temperature of the measuring point is simplified to be the average temperature of the same day, and the additional error delta f of the same day temperature is obtained (T) ;
T vp =T v -(y TEM (v)×Δf (T) ) (4)
S vp =S v -(S TEM (v)×Δf (T) ) (5)
Respectively calculating each time T vp And form sequence data T p Respectively calculate S vp And form sequence data S p 。
Further, the specific steps of S3 are as follows:
s301, drawing time sequence T p 、S p Respectively determining T through characteristic judgment of the two graphs p 、S p Selecting a model and model parameters;
s302 according to T p 、S p The conforming model selection result and model parameter selection result are used for importing a time sequence into a model to obtain prediction models M1 and M2, and the prediction models are used for predicting to obtain T r 、S r 、T rv 、S rv ;
S303, predicting the frequency FRE of the time period to be detected by adopting the same arima model algorithm r HUM of humidity r Load LOA r Vibration VIB r Magnetic field MAG r For historical frequency, humidity, load, vibration, magnetic field data, and R v Leading the frequency FRE into an LSTM long-term memory neural network, obtaining a residual component prediction model M3 through training, and predicting the frequency FRE for the future period of the transformer to be tested r 、HUM r 、LOA r 、VIB r 、MAG r Leading in a trained model M3 to obtain R rv And R is r 。
Further, the specific steps of S4 are as follows:
according to weather forecast of weather department to day to be measured, obtaining temperature data of each moment of day to be measured, and recording difference value of temperature data and 20 ℃ as delta t p And calculates an additional error Δf according to equation (3) (T)p Adding the trend and the period prediction result of the stripping temperature factors to the error of the temperature influence of the day to be detected and the residual prediction result output by the neural network to obtain the final prediction result, and finally obtaining the error f predicted by the transformer at the moment v rv The method comprises the following steps:
f rv =T rv +S rv +Δf (T)p +R rv
calculate each time f rv And forms an error prediction sequence F of a transformer for a predicted time r 。
The beneficial effects of the invention are as follows: accurately predicting the error of the current transformer in the future period, and timely finding out the transformer to be out of tolerance so as to realize risk early warning of the current transformer; taking an error evaluation value obtained by an error evaluation algorithm as basic training data of the text, decomposing an error sequence according to an error tracing result of the current transformer and combining the structural characteristics of the current transformer, and respectively correcting and then respectively predicting;
the decomposition and combination of the error sequence data of the transformer; an influence algorithm of temperature on the trend component and the period component; the trend component and the period component are predicted after the influence of the peeling temperature is greatly influenced, so that trend and period prediction is focused on the reasons of temperature accidents, and the accuracy is improved; the effect of various factors on the residual component is taken into account.
Drawings
FIG. 1 is a flow chart of a method for predicting a metering error value of a current transformer according to the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.
A method for predicting metering error value of a current transformer comprises the following steps:
s1, taking an error evaluation value obtained by a mutual inductor error evaluation algorithm as historical training data, and decomposing a time sequence by using an STL method to obtain three components;
s2, collecting data of various influencing factors, calculating influences of temperature on trend components and periodic components, and obtaining trend components and periodic components after the influence of the stripping temperature for subsequent prediction;
s3, predicting trend components and periodic components of the future time interval to be measured after stripping temperature influence factors by adopting an arima model algorithm, adding an additional error corresponding to the predicted temperature of the time interval to be measured as a trend component and periodic component prediction result, and introducing the predicted influence factors into an LSTM neural network model to predict future time interval residual error components;
s4, combining the error components of the separate prediction to generate a future transformer error prediction result.
The S1 specifically comprises the following steps: taking an error estimation value obtained by a mutual inductor error estimation algorithm as historical training data, wherein the acquisition frequency is 1 per hour, the data quantity is N, a predicted data sequence is formed, and the predicted data sequence is recorded as: y is Y V ;
The STL method is adopted to decompose the time corresponding to the collected historical transformer error sequence data into three components, namely a trend component, a period component and a residual error component, namely:
Y v =T v +S v +R v (1)
wherein T is v 、S v 、R v The trend component, the periodic component and the residual component of the v moment are respectively.
The specific implementation process of the STL method is as follows:
TV (k), SV (k) are assumed to be trend component, period component, TV (k) =0 at the end of the kth-1 st time in the inner loop.
Trending: subtracting the trend component Yv' =yv-Tv (k) of the previous round of results;
periodic subsequence smoothing: regression is performed on each subsequence by using the LOESS (q=nn(s), d=1) (the LOESS refers to local weighted regression, n(s) is a LOESS smoothing parameter, q is a data subset acquisition range, and d is an order in local regression), and the regression is extended forward and backward for one period respectively; the smoothed results form a temporal periodic time series, denoted Cv (k+1), v= -N (p) +1, -n+n (p);
low-flux filtering of periodic subsequences: the resulting sequence Cv (k+1) of the previous step is sequentially subjected to a running average of lengths n (p) n (p), 3 (n (p) is the number of samples of one cycle, which is 24), and then to a log (q=n n(1) D=1) regression (n(s) is the LOESS smoothing parameter) to obtain the result sequence
Removing the smooth periodic subsequence trend:
and (3) cycle removal: the periodic component is subtracted from the periodic component,
trend smoothing: for the sequence after the removal period, a leave is made (q=n n(t) D=1) regression to obtain trend components
The trend component time sequence, the period component time sequence and the residual component time sequence after the mutual inductor error time sequence decomposition are obtained through the steps and are marked as T, S, R, wherein the S period is 24.
The specific steps of the S2 are as follows:
s201, collecting time data corresponding to each influence factor, wherein the time data is once per hour;
in one specific embodiment, the method comprises the steps of environmental temperature, environmental humidity with seasonal trend change and daily period change, and a plurality of error influence factors with high randomness such as primary frequency, secondary load, body vibration and environmental magnetic field.
S202, recording a temperature time sequence of acquired contemporaneous historical training data as TEM, a humidity time sequence as HUM, a primary frequency time sequence as FRE, a load time sequence as LOA, a VA, a vibration amplitude time sequence as VIB, a mm and a magnetic field time sequence as MAG;
s203, fitting the TEM by using a least square method to obtain a fitting curve y TEM (v)=a 0 +a 1 x+...+a n x n The fitted n-th-order item curve; taking n=3 as an influence factor of temperature on T;
s204, recording the calculated average value of the data points of the historical time series data TEM at 6 days asThe average value of the data points at 12 daily in TEM was calculated and recorded as +.>Taking 24h as a period, and obtaining a fitting curve S of the temperature under the influence of the period according to the lowest temperature at 6 am and the highest temperature at 12 am TEM (v):
Wherein V' is the time of the day of V;
s205, calculating trend component T and periodic component S respectively, stripping temperature factor and then predicting trend component T p Periodic component S p The temperature additional error formula is:
wherein B is rated load;is a preset constant; a, a c Is the temperature coefficient; Δt is the difference between the temperature of the measuring point and 20 ℃; omega n Is the rated angular frequency; i 1 Is the primary current; the temperature of the measuring point is simplified to be the average temperature of the same day, and the additional error delta f of the same day temperature is obtained (T) ;
T vp =T v -(y TEM (v)×Δf (T) ) (4)
S vp =S v -(S TEM (v)×Δf (T) ) (5)
Respectively calculating each time T vp And form sequence data T p Respectively calculate S vp And form sequence data S p 。
The specific steps of the S3 are as follows:
s301, drawing time sequence T p 、S p Respectively determining T through characteristic judgment of the two graphs p 、S p Selecting a model and model parameters;
in a specific embodiment, the partial autocorrelation coefficients should be zero after the p-th order, as viewed in the figure, and they are said to have tail-biting, and the autocorrelation coefficients cannot be zero after a certain step (tail-biting) but are exponentially decaying (or in the form of a sine wave), as they have tail-biting
S302 according to T p 、S p The conforming model selection result and model parameter selection result are used for importing a time sequence into a model to obtain prediction models M1 and M2, and the prediction models are used for predicting to obtain T r 、S r 、T rv 、S rv ;
S303, predicting the frequency FRE of the time period to be detected by adopting the same arima model algorithm r HUM of humidity r Load LOA r Vibration VIB r Magnetic field MAG r For historical frequency, humidity, load, vibration, magnetic field data, and R v Leading the frequency FRE into an LSTM long-term memory neural network, obtaining a residual component prediction model M3 through training, and predicting the frequency FRE for the future period of the transformer to be tested r 、HUM r 、LOA r 、VIB r 、MAG r Leading in a trained model M3 to obtain R rv And R is r 。
The specific steps of the S4 are as follows:
according to weather forecast of weather department to day to be measured, obtaining temperature data of each moment of day to be measured, and recording difference value of temperature data and 20 ℃ as delta t p And calculates an additional error Δf according to equation (3) (T)p Adding the trend and the period prediction result of the stripping temperature factors to the error of the temperature influence of the day to be detected and the residual prediction result output by the neural network to obtain the final prediction result, and finally obtaining the error f predicted by the transformer at the moment v rv The method comprises the following steps:
f rv =T rv +S rv +Δf (T)p +R rv
calculate each time f rv And forms an error prediction sequence F of a transformer for a predicted time r 。
And the error of the current transformer in the future period is accurately predicted, the transformer to be out of tolerance is timely found, and the risk early warning of the current transformer is realized.
The foregoing examples merely illustrate embodiments of the invention and are described in more detail and are not to be construed as limiting the scope of the invention. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the invention, which are all within the scope of the invention. Accordingly, the scope of protection of the present patent is to be determined by the appended claims.
Claims (5)
1. The method for predicting the metering error value of the current transformer is characterized by comprising the following steps of:
s1, taking an error evaluation value obtained by a mutual inductor error evaluation algorithm as historical training data, and decomposing a time sequence by using an STL method to obtain three components;
s2, collecting data of various influencing factors, calculating influences of temperature on trend components and periodic components, and obtaining trend components and periodic components after the influence of the stripping temperature for subsequent prediction;
s3, predicting trend components and periodic components of the future time interval to be measured after stripping temperature influence factors by adopting an arima model algorithm, adding an additional error corresponding to the predicted temperature of the time interval to be measured as a trend component and periodic component prediction result, and introducing the predicted influence factors into an LSTM neural network model to predict future time interval residual error components;
s4, combining the error components of the separate prediction to generate a future transformer error prediction result.
2. The method for predicting a metering error value of a current transformer according to claim 1, wherein S1 specifically is: taking an error estimation value obtained by a mutual inductor error estimation algorithm as historical training data, wherein the acquisition frequency is 1 per hour, the data quantity is N, a predicted data sequence is formed, and the predicted data sequence is recorded as: y is Y V ;
The STL method is adopted to decompose the time corresponding to the collected historical transformer error sequence data into three components, namely a trend component, a period component and a residual error component, namely:
Y v =T v +S v +R v (1)
wherein T is v 、S v 、R v The trend component, the periodic component and the residual component of the v moment are respectively.
3. The method for predicting the metering error value of the current transformer according to claim 2, wherein the specific steps of S2 are as follows:
s201, collecting time data corresponding to each influence factor, wherein the time data is once per hour;
s202, recording a temperature time sequence of acquired contemporaneous historical training data as TEM, a humidity time sequence as HUM, a primary frequency time sequence as FRE, a load time sequence as LOA, a VA, a vibration amplitude time sequence as VIB, a mm and a magnetic field time sequence as MAG;
S203fitting TEM by least square method to obtain a fitted curve, y TEM (v)=a 0 +a 1 x+...+a n x n The fitted n-th-order item curve; taking n=3 as an influence factor of temperature on T;
s204, recording the calculated average value of the data points of the historical time series data TEM at 6 days asThe average value of the data points at 12 daily in TEM was calculated and recorded as +.>Taking 24h as a period, and obtaining a fitting curve S of the temperature under the influence of the period according to the lowest temperature at 6 am and the highest temperature at 12 am TEM (v):
Wherein V' is the time of the day of V;
s205, calculating trend component T and periodic component S respectively, stripping temperature factor and then predicting trend component T p Periodic component S p The temperature additional error formula is:
wherein B is rated load;is a preset constant; a, a c Is the temperature coefficient; Δt is the difference between the temperature of the measuring point and 20 ℃; omega n Is the rated angular frequency; i 1 Is the primary current; the temperature of the measuring point is simplified to be the average temperature of the same day, and the additional error delta f of the same day temperature is obtained (T) ;
T vp =T v -(y TEM (v)×Δf (T) ) (4)
S vp =S v -(S TEM (v)×Δf (T) ) (5)
Respectively calculating each time T vp And form sequence data T p Respectively calculate S vp And form sequence data S p 。
4. The method for predicting a metering error value of a current transformer according to claim 3, wherein the specific step of S3 is as follows:
s301, drawing time sequence T p 、S p Respectively determining T through characteristic judgment of the two graphs p 、S p Selecting a model and model parameters;
s302 according to T p 、S p The conforming model selection result and model parameter selection result are used for importing a time sequence into a model to obtain prediction models M1 and M2, and the prediction models are used for predicting to obtain T r 、S r 、T rv 、S rv ;
S303, predicting the frequency FRE of the time period to be detected by adopting an arima model algorithm r HUM of humidity r Load LOA r Vibration VIB r Magnetic field MAG r First, as model inputs, historical frequency, humidity, load, vibration, magnetic field data, and historical R v As output, the residual component prediction model M3 is obtained by training after being led into an LSTM long-term memory neural network, and the predicted frequency FRE is used for the future period of the transformer to be tested r 、HUM r 、LOA r 、VIB r 、MAG r Leading in a trained model M3 to obtain R rv And R is r 。
5. The method for predicting a metering error value of a current transformer according to claim 4, wherein the specific step of S4 is as follows:
according to weather forecast of weather department to day to be measured, obtaining temperature data of each moment of day to be measured, and recording difference value of temperature data and temperature data of 20 DEG CMake Δt p And calculates an additional error Δf according to equation (3) (T)p Adding the trend and the period prediction result of the stripping temperature factors to the error of the temperature influence of the day to be detected and the residual prediction result output by the neural network to obtain the final prediction result, and finally obtaining the error f predicted by the transformer at the moment v rv The method comprises the following steps:
f rv =T rv +S rv +Δf (T)p +R rv
calculate each time f rv And forms an error prediction sequence F of a transformer for a predicted time r 。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310699195.2A CN116720622A (en) | 2023-06-14 | 2023-06-14 | Method for predicting metering error value of current transformer |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202310699195.2A CN116720622A (en) | 2023-06-14 | 2023-06-14 | Method for predicting metering error value of current transformer |
Publications (1)
Publication Number | Publication Date |
---|---|
CN116720622A true CN116720622A (en) | 2023-09-08 |
Family
ID=87869311
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202310699195.2A Pending CN116720622A (en) | 2023-06-14 | 2023-06-14 | Method for predicting metering error value of current transformer |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN116720622A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117233687A (en) * | 2023-11-13 | 2023-12-15 | 威胜集团有限公司 | CVT initial error assessment method, medium and terminal based on historical data |
CN117434486A (en) * | 2023-12-20 | 2024-01-23 | 智联信通科技股份有限公司 | DC shunt metering error analysis processing method |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5559431A (en) * | 1993-10-08 | 1996-09-24 | Micro-Epsilon Messtechnik Gmbh & Co. Kg | Method of calibrating a sensor |
CN114692505A (en) * | 2022-04-12 | 2022-07-01 | 广东电网有限责任公司 | Error prediction method and device for capacitor voltage transformer |
CN115438576A (en) * | 2022-08-18 | 2022-12-06 | 三峡大学 | Electronic voltage transformer error prediction method based on Prophet, self-attention mechanism and time series convolution network |
CN116049629A (en) * | 2023-03-29 | 2023-05-02 | 国网福建省电力有限公司 | Voltage transformer error state prediction method, system, equipment and medium |
-
2023
- 2023-06-14 CN CN202310699195.2A patent/CN116720622A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5559431A (en) * | 1993-10-08 | 1996-09-24 | Micro-Epsilon Messtechnik Gmbh & Co. Kg | Method of calibrating a sensor |
CN114692505A (en) * | 2022-04-12 | 2022-07-01 | 广东电网有限责任公司 | Error prediction method and device for capacitor voltage transformer |
CN115438576A (en) * | 2022-08-18 | 2022-12-06 | 三峡大学 | Electronic voltage transformer error prediction method based on Prophet, self-attention mechanism and time series convolution network |
CN116049629A (en) * | 2023-03-29 | 2023-05-02 | 国网福建省电力有限公司 | Voltage transformer error state prediction method, system, equipment and medium |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN117233687A (en) * | 2023-11-13 | 2023-12-15 | 威胜集团有限公司 | CVT initial error assessment method, medium and terminal based on historical data |
CN117233687B (en) * | 2023-11-13 | 2024-02-06 | 威胜集团有限公司 | CVT initial error assessment method, medium and terminal based on historical data |
CN117434486A (en) * | 2023-12-20 | 2024-01-23 | 智联信通科技股份有限公司 | DC shunt metering error analysis processing method |
CN117434486B (en) * | 2023-12-20 | 2024-03-08 | 智联信通科技股份有限公司 | DC shunt metering error analysis processing method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN116720622A (en) | Method for predicting metering error value of current transformer | |
CN106533750B (en) | The forecasting system and method for non-stationary type application user concurrent amount under a kind of cloud environment | |
CN106505557B (en) | Remote measurement error identification method and device | |
CN109726865A (en) | Customer charge probability density prediction technique, device and storage medium based on EMD-QRF | |
CN113312413B (en) | Enterprise carbon emission abnormity monitoring method based on electric power big data | |
JP2009294969A (en) | Demand forecast method and demand forecast device | |
Liu et al. | Real-time corporate carbon footprint estimation methodology based on appliance identification | |
CN109284933B (en) | Electronic transformer state evaluation system and method based on mathematical statistics | |
Lv et al. | Short-term wind speed forecasting based on non-stationary time series analysis and ARCH model | |
CN102422235B (en) | Determining an equivalent mechanical load | |
CN116306798A (en) | Ultra-short time wind speed prediction method and system | |
CN116933044A (en) | Intelligent processing method and system for power supply data | |
Tan et al. | Intelligent online health estimation for lithium-ion batteries based on a parallel attention network combining multivariate time series | |
CN115508770A (en) | KL-NB algorithm-based electric energy meter operation state online evaluation method | |
CN117290685A (en) | Power plant power equipment expert diagnosis system and method based on historical data | |
CN111192163B (en) | Generator reliability medium-short term prediction method based on wind turbine generator operating data | |
Xin et al. | Short-term load forecasting for electric vehicle charging stations based on time series distance measuring | |
CN116320824A (en) | Thing allies oneself with electric energy meter of electrified energy quality module | |
Ranasingh et al. | Calibration and frequency estimation in sensors for electrical parameter measurement using regression and metaheuristic based models | |
CN113935523A (en) | Voltage trend prediction method based on voltage change rule | |
JP2016086519A (en) | Nonlinear prediction method and device of power consumption | |
CN110472801B (en) | Electromagnetic environment assessment method and system for direct-current transmission line | |
CN114492923A (en) | Long-time scale power prediction method | |
Li et al. | Extraction of Abnormal Points from On-line Operation Data of Intelligent Meter Based on LSTM | |
CN112749820A (en) | Wind power prediction method and system |
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 |