A power transmission line overhead ground wire power loss prediction method, device and medium

Abstract

The application discloses a power transmission line overhead ground wire electric energy loss prediction method and device and a medium, and relates to the technical field of electric power engineering. The method comprises the following steps: obtaining historical load data and historical weather feature data of a target power transmission line overhead ground wire, and preprocessing the historical weather feature data and date feature data; screening target weather feature data; obtaining prediction weather feature data meeting second preset requirements, and applying a clustering algorithm; inputting the prediction weather feature data cluster set and the date feature data cluster set into a load prediction model to obtain prediction load data; considering the change relationship between the electric conductance value of an insulator and precipitation to construct an initial electric energy loss accurate prediction model; inputting the prediction weather feature data, the prediction load data and target power transmission line overhead ground wire parameter data into an electric energy loss prediction model to obtain electric energy loss data of the target power transmission line overhead ground wire; and combining actual measurement data to correct the prediction model to obtain an electric energy loss prediction model. The application can improve prediction accuracy.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power engineering, in particular to a power loss prediction method and device for overhead ground wire of power transmission line and medium. BACKGROUND

[0002] The overhead ground wire is the most basic lightning protection measure of the power transmission line, and through the shunt effect of lightning current, the lightning current intensity entering the grounding tower can be effectively reduced, and then the tower top potential can be reduced, which has very important significance for the safe and stable operation of the power grid, but with the improvement of the voltage grade of the power transmission line, the power loss of the overhead ground wire also becomes larger and larger.

[0003] At present, the OPGW of high-voltage power transmission line basically adopts the mode of grounding at each base, and the ordinary lightning conductor mostly adopts the mode of sectional insulation and grounding at one point. The OPGW grounded at each tower forms a loop with the ground, and the induced current causes huge power loss. Due to the rapid development of industry, a large amount of electric power resources is wasted unknowingly. The traditional prediction method mainly aims at the prediction research of the overhead power transmission line structure, and fails to consider the influence of precipitation on the conductance value of the overhead ground wire and the influence of load on the electromagnetic induction voltage and electrostatic induction voltage of the overhead ground wire, so that the prediction of the power loss of the overhead ground wire is not accurate enough in precision. SUMMARY

[0004] The purpose of the present application is to provide a power loss prediction method and device for overhead ground wire of power transmission line and medium, which can improve the prediction precision of the power loss of the overhead ground wire, reflect the power loss value of the OPGW optical cable line, and promote the precision of the insulation transformation of the OPGW.

[0005] To achieve the above purpose, the present application provides the following scheme:

[0006] A power loss prediction method for overhead ground wire of power transmission line, the power loss prediction method comprising:

[0007] Obtaining historical load data of a target overhead ground wire of a target power transmission line and historical weather feature data and corresponding date feature data in a target area where the target overhead ground wire is located; the historical weather feature data comprises historical temperature data, historical precipitation data, historical humidity data, historical air pressure data and historical wind speed data; and the date feature data comprises weekdays and non-weekdays.

[0008] Pretreating the historical load data and the historical weather feature data to obtain optimized historical load data and historical weather feature data; the pretreatment operation comprises eliminating abnormal values and correcting;

[0009] Filter target weather feature data from the optimized historical weather feature data; the correlation degree of the target weather feature data with the optimized historical load data meets a first preset requirement;

[0010] Obtain predicted weather feature data meeting a second preset requirement, and apply a clustering algorithm to the predicted weather feature data meeting the second preset requirement and corresponding date feature data to obtain a predicted weather feature data cluster set and a date feature data cluster set; the predicted weather feature data includes predicted temperature data, predicted precipitation data, predicted humidity data, predicted air pressure data, and predicted wind speed data;

[0011] Input the predicted weather feature data cluster set and the date feature data cluster set into a load prediction model to obtain predicted load data; the load prediction model is obtained by training an improved grey prediction model using a load prediction training data set; the load prediction training data set includes historical weather feature data cluster set samples, corresponding date feature data cluster set samples, and corresponding load data samples; the historical weather feature data cluster set samples are obtained by applying a clustering algorithm to the target weather feature data;

[0012] According to the relationship between the insulator conductance value and the precipitation, and the relationship between the overhead ground wire electromagnetic induction voltage value and the static induction voltage value and the load, an initial electric energy loss accurate prediction model is constructed;

[0013] Input the predicted weather feature data, the predicted load data, and the target overhead ground wire parameter data of the transmission line into an electric energy loss prediction model to obtain the electric energy loss data of the target overhead ground wire of the transmission line; the electric energy loss prediction model is obtained by training the initial electric energy loss accurate prediction model using an electric energy loss prediction training data set; the electric energy loss prediction training data set includes the historical weather feature data, the historical load data, the target overhead ground wire parameter data of the transmission line, and corresponding historical electric energy loss data of the target overhead ground wire of the transmission line; the target overhead ground wire parameter data of the transmission line includes the overhead ground wire electromagnetic induction voltage value, the static induction voltage value, and the insulator conductance value.

[0014] Optionally, the historical load data and the historical weather feature data are preprocessed to obtain optimized historical load data and historical weather feature data, specifically including:

[0015] Perform outlier detection on the historical load data and the historical weather feature data;

[0016] Eliminate the outliers from the historical load data and the historical weather feature data;

[0017] The mean value algorithm is applied to the historical load data and the historical weather feature data after removing the abnormal values to obtain optimized historical load data and historical weather feature data.

[0018] Optionally, 3-sigma abnormal value detection is performed on the historical load data and the historical weather feature data.

[0019] Optionally, target weather feature data is screened from the historical weather feature data, and specifically includes:

[0020] The Spearman correlation coefficient method is applied to calculate the correlation coefficient of the historical weather feature data and the historical load data.

[0021] According to the correlation coefficient, the correlation degree of the historical weather feature data and the historical load data is determined; the correlation degree includes strong correlation, weak correlation and no correlation; when the correlation coefficient is greater than or equal to a first preset threshold, the correlation degree is strong correlation; when the correlation coefficient is less than the first preset threshold and greater than a second preset threshold, the correlation degree is weak correlation; when the correlation coefficient is less than or equal to the second preset threshold, the correlation degree is no correlation.

[0022] It is determined whether the correlation degree is the strong correlation and the weak correlation.

[0023] When the correlation degree is the strong correlation and the weak correlation, the corresponding historical weather feature data is target weather feature data.

[0024] Optionally, predicted weather feature data that meets a second preset requirement is obtained, and a clustering algorithm is applied to the predicted weather feature data and the date feature data to obtain predicted weather feature data clusters and date feature data clusters, and specifically includes:

[0025] Initial weather feature data of each meteorological monitoring station in the target area at each preset time in a preset time period is obtained.

[0026] Initial weather feature data that meets a second preset requirement in terms of the correlation degree with the historical load data is screened from the initial weather feature data to obtain screened weather feature data.

[0027] The average value of the screened weather feature data is calculated to obtain weather feature data of each preset time in the target area, and the weather feature data of each preset time in the target area is taken as predicted weather feature data in the preset time period.

[0028] The k-means clustering algorithm is applied to process the predicted weather feature data and the corresponding date feature data to obtain predicted weather feature data clusters and date feature data clusters.

[0029] Optionally, the process of constructing the initial electric energy loss accurate prediction model specifically comprises:

[0030] According to the principle of electromagnetic coupling, a first relationship is constructed, and an electromagnetic induction voltage of the overhead ground wire of the transmission line is calculated according to the first relationship; the first relationship is a relationship between the electromagnetic induction voltage of the overhead ground wire of the transmission line and a first parameter of the transmission line; the first parameter of the transmission line includes a total number of towers of the transmission line, predicted load data, line voltage, power factor, rotation factor, spacing between the overhead ground wire and the conductor in a base tower, and an electromagnetic induction voltage correction factor;

[0031] According to the principle of electrostatic induction, a second relationship is constructed, and an electrostatic induction voltage of the overhead ground wire of the transmission line is calculated according to the second relationship; the second relationship is a relationship between the electrostatic induction voltage of the overhead ground wire of the transmission line and a second parameter of the transmission line; the second parameter of the transmission line includes a charge coefficient of the conductor, a charge on the conductor, a charge coefficient on the overhead ground wire, a charge on the overhead ground wire in the base tower, and an electrostatic induction voltage correction factor;

[0032] According to the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, and a first state of a first classification, a third relationship is determined, and a first electric energy loss of the overhead ground wire in a tower-by-tower grounding mode is calculated; the first classification is an overhead ground wire of the transmission line in the tower-by-tower grounding mode; the first state is a non-idling state of the transmission line; the third relationship is a relationship between the first electric energy loss of the overhead ground wire in the tower-by-tower grounding mode and a third parameter of the transmission line; the third parameter of the transmission line includes the electromagnetic induction voltage of the overhead ground wire, the electrostatic induction voltage of the overhead ground wire, self-impedance of the overhead ground wire, mutual impedance of the overhead ground wire, and grounding resistance of a tower of the transmission line;

[0033] According to a second state of the first classification, a fourth relationship is determined, and a second electric energy loss of the overhead ground wire in the tower-by-tower grounding mode is calculated according to the fourth relationship; the second state is an idling state of the transmission line; the fourth relationship is a relationship between the second electric energy loss of the overhead ground wire in the tower-by-tower grounding mode and a fourth parameter of the transmission line; the fourth parameter of the transmission line includes a charge on the overhead ground wire in the base tower, an angular frequency, and grounding resistance of a base tower of the transmission line;

[0034] According to the first state of the second classification, a fifth relationship is determined, and a conductance of the overhead ground wire insulator in the sectional insulation mode is calculated according to the fifth relationship; the second classification is the overhead ground wire of the transmission line in the sectional insulation mode; the fifth relationship is a relationship between the conductance of the overhead ground wire insulator in the sectional insulation mode and a fifth parameter of the transmission line; and the fifth parameter includes an effective area of an insulator surface, a relative humidity of weather, a rainfall intensity, and a surface conductivity variation coefficient.

[0035] According to the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, and the second state of the second classification, a sixth relationship is determined, and an electrical energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode in a non-idling state of the transmission line is calculated according to the sixth relationship; the sixth relationship is a relationship between the electrical energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode in the non-idling state of the transmission line and a sixth parameter; and the sixth parameter includes the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, a conductance of a ground wire insulator affected by the precipitation, a self impedance of the overhead ground wire, and a sectional number.

[0036] According to the second state of the second classification, a seventh relationship is determined, and an electrical energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode in an idling state of the transmission line is calculated according to the seventh relationship; the seventh relationship is a relationship between the electrical energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode in the idling state of the transmission line and a seventh parameter; and the seventh parameter includes the electrostatic induction voltage of the overhead ground wire of the transmission line and the conductance of the ground wire insulator affected by the precipitation.

[0037] According to the predicted load data and the predicted precipitation data meeting the second preset requirement, an eighth relationship is determined, and one of candidate models is selected as the initial electrical energy loss accurate prediction model according to the eighth relationship; the candidate models include the fourth relationship, the fifth relationship, the sixth relationship, the seventh relationship, and 0.

[0038] Optionally, the electrical energy loss prediction method further includes:

[0039] According to the electrical energy loss data of the target overhead ground wire of the transmission line output by the electrical energy loss prediction model and the actual electrical energy loss data of the target overhead ground wire of the transmission line, an accuracy of the electrical energy loss prediction model is determined.

[0040] Optionally, the electrical energy loss prediction method further includes:

[0041] According to the predicted electrical energy loss data output by the initial electrical energy loss accurate prediction model and the corresponding measured electrical energy loss data, the initial electrical energy loss accurate prediction model is corrected.

[0042] A computer device comprises a memory, a processor to store a computer program on the memory and run the computer program on the processor, and the processor executes the computer program to implement the power loss prediction method of the overhead ground wire of the power transmission line according to any one of the above.

[0043] A computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the power loss prediction method of the overhead ground wire of the power transmission line according to any one of the above.

[0044] According to the specific embodiments of the present application, the following technical effects are disclosed:

[0045] The present application provides a power loss prediction method, device and medium for the overhead ground wire of the power transmission line, which considers the influence of precipitation on the electric conductivity value and the influence of load on the electromagnetic induction and electrostatic induction voltage value to construct a power loss prediction model, and combines the actual measurement data to correct the prediction model by calculating the residual error and the correction factor, so as to improve the prediction accuracy, reflect the power loss value of the OPGW optical cable line, and promote the accuracy of the OPGW insulation reconstruction. BRIEF DESCRIPTION OF DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0047] Figure 1 The power loss prediction method flowchart of the overhead ground wire of the power transmission line is provided for Embodiment 1 of the present application.

[0048] Figure 2 It is a normal distribution curve diagram of 3-sigma model.

[0049] Figure 3 It is an internal structure diagram of a computer device. DETAILED DESCRIPTION

[0050] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0051] The application aims to provide a power loss prediction method for overhead ground wires of power transmission lines, a device and a medium, and aims to improve the prediction accuracy of power loss of overhead ground wires of power transmission lines.

[0052] In order to make the above-mentioned purposes, characteristics and advantages of the application more obvious and easy to understand, the application will be further described in detail below in combination with the drawings and specific embodiments.

[0053] Embodiment 1

[0054] As shown in the figure, the power loss prediction method for overhead ground wires of power transmission lines in this embodiment comprises: Figure 1

[0055] Step S1: obtaining historical load data of a target overhead ground wire of a power transmission line and historical weather feature data and corresponding date feature data in a target area where the target overhead ground wire of the power transmission line is located; the historical weather feature data comprises historical temperature data, historical precipitation data, historical humidity data, historical air pressure data and historical wind speed data; the date feature data comprises weekdays and non-weekdays.

[0056] Every week, the load data will be obviously different due to the life and work of residents on weekdays and non-weekdays.

[0057] In the subsequent steps, the temperature, precipitation, humidity, air pressure and wind speed data are first selected by the 3-sigma algorithm, and then the weather data meeting the 3-sigma algorithm selection is used to predict the load by using the clustering algorithm and the improved grey prediction algorithm.

[0058] Step S2: preprocessing the historical load data and the historical weather feature data to obtain optimized historical load data and historical weather feature data; the preprocessing operation comprises outlier rejection and correction.

[0059] S2 specifically comprises:

[0060] Step S21: performing outlier detection on the historical load data and the historical weather feature data.

[0061] After the historical load data and the historical weather feature data are collected, there may be a situation that a certain data is very high or very low, in order to make the historical load data and the historical weather feature data more meaningful, it is necessary to reject and clean the abnormal data in the historical load data and the historical weather feature data.

[0062] Specifically, 3-sigma outlier detection is performed on the historical load data and the historical weather feature data.

[0063] ​Step S22: removing the outliers from the historical load data and the historical weather feature data.

[0064] Step S23: applying a mean value algorithm to correct the mean value of the historical load data and the historical weather feature data after removing the outliers, to obtain optimized historical load data and historical weather feature data.

[0065] In practical application, 3-sigma outlier detection is performed on historical load data and historical weather feature data respectively. The mean value and standard deviation σ of historical data are calculated to determine whether each value in the data is within the interval [μ-3σ, μ+3σ], as shown in formula (1), to obtain the maximum threshold and minimum threshold of the data within the historical time range. The maximum threshold is the mean value of the historical data plus three times the standard deviation, and the minimum threshold is the mean value of the historical data minus three times the standard deviation. If it is not within the threshold range, it is marked as an outlier. Then, the mean value method is used to replace the outliers by taking the average value of the historical data corresponding to the time points before and after the outliers, to obtain the optimized historical load data and historical weather feature data. Figure 2

[0066] Step S3: selecting target weather feature data from the optimized historical weather feature data; the target weather feature data is the historical weather feature data whose correlation degree with the historical load data meets a first preset requirement.

[0067] S3 specifically includes:

[0068] Step S31: applying a Spearman correlation coefficient method to calculate the correlation coefficient of the historical weather feature data and the historical load data.

[0069] Step S32: determining the correlation degree of the historical weather feature data and the historical load data according to the correlation coefficient; the correlation degree includes strong correlation, weak correlation and no correlation; when the correlation coefficient is greater than or equal to a first preset threshold, the correlation degree is strong correlation; when the correlation coefficient is less than the first preset threshold and greater than a second preset threshold, the correlation degree is weak correlation; when the correlation coefficient is less than or equal to the second preset threshold, the correlation degree is no correlation.

[0070] Step S33: determining whether the correlation degree is the strong correlation and the weak correlation.

[0071] Step S34: when the correlation degree is the strong correlation and the weak correlation, the corresponding historical weather feature data is target weather feature data.

[0072] In practical application, the Spearman correlation coefficient method is used to calculate the correlation of each historical weather data and each historical load data, r​s The calculation is as follows:

[0073]

[0074] In the formula, ψ is the rank number, d i is the rank difference between u and y i , u is historical load data, y i is one of historical temperature data, historical precipitation data, historical humidity data, historical air pressure data, and historical wind speed data.

[0075] According to the size of the Spearman correlation coefficient, the judgment rule is:

[0076] When |r s |≥0.7, it is judged as strong correlation, when 0.2 s |<0.7, it is judged as weak correlation, and when r s |≤0.2, it is judged as not related.

[0077] Step S4: obtaining predicted weather feature data meeting the second preset requirement, and applying a clustering algorithm to the predicted weather feature data meeting the second preset requirement and the corresponding date feature data to obtain a predicted weather feature data cluster set and a date feature data cluster set.

[0078] S4 specifically includes:

[0079] Step S41: obtaining initial weather feature data of each meteorological monitoring station in the target area at each preset time in a preset time period.

[0080] Step S42: selecting initial weather feature data meeting a second preset requirement in terms of correlation degree with the historical load data from the initial weather feature data to obtain screened weather feature data.

[0081] Step S43: calculating the average value of the screened weather feature data to obtain weather feature data of each preset time in the target area, and taking the weather feature data of each preset time in the target area as predicted weather feature data in the preset time period.

[0082] Step S44: processing the predicted weather feature data and the corresponding date feature data using a k-means clustering algorithm to obtain a predicted weather feature data cluster set and a date feature data cluster set.

[0083] In actual application, weather feature data of multiple meteorological monitoring stations in the target area that is strongly and weakly correlated with historical load data in a preset time is collected, and the average value is calculated.

[0084] Any n initial points in the data set are selected as clustering centers, for each data point in the predicted weather feature data and the date feature data, the distance from each center point is calculated respectively, and it is assigned to the nearest clustering center, for each cluster, the mean of all data points is calculated, which is the new center, repeat the above steps until the new center no longer changes or meets the predetermined number of iterations, to obtain the predicted weather feature data cluster set and the date feature data cluster set.

[0085] Step S5: inputting the predicted weather feature data cluster set and the date feature data cluster set into a load prediction model to obtain predicted load data; the load prediction model is obtained by training an improved grey prediction model through a load prediction training data set; the load prediction training data set includes a historical weather feature data cluster set sample, a corresponding date feature data cluster set sample and a corresponding load data sample; the historical weather feature data cluster set sample is obtained by applying a clustering algorithm to the target weather feature data; the corresponding date feature data cluster set sample is obtained by applying a clustering algorithm to the date feature data corresponding to the target weather feature data.

[0086] The predicted weather feature data cluster set and the date feature data cluster set are input into the improved grey prediction model to obtain predicted load data, and the improved grey prediction model is corrected through residual calculation combined with the actually measured load data.

[0087] In practical application, grey prediction mainly uses GM model to analyze and predict the evolution trend and law of system behavior characteristics, and can estimate the time point of abnormal state of system behavior. It is a prediction method that establishes a mathematical model and makes a prediction through a small amount of incomplete information, and is widely used in various fields, especially can make prediction analysis on future development trend according to the existing data.

[0088] When processing some data sequences with insufficient smoothness, the accuracy of the grey model will be greatly reduced. In this paper, the original data is first optimized by an exponential function to improve its smoothness and improve the prediction accuracy.

[0089] Specifically, the reference data is set as: x (0) = (x (0) (1), x (0) (2), …, x (0) (n)).

[0090] First, the original data is optimized by an exponential function as follows:

[0091] y (0) (k) = e x(0)(k)k = 1, 2,.., n; wherein e is the base; the original data is reference data. The new data set generated is transformed according to the GM(1, 1) model to obtain a new response equation, specifically, through an exponential function, the new response sequence is: y (0) (k+1) * = y (1) (k+1) * -y (1) (k) * .

[0092] The restored value of the original sequence is: x (0) (k) * = ln(y (0) (k) * ).

[0093] By accumulating the generated sequence, an improved grey differential equation is established: x (0) (k) + λx (1) (k) = ζ, wherein λ is a development coefficient and ζ is a grey action amount; the least square method is used to solve the estimated values of λ and ζ, and the predicted value of the grey prediction model is:

[0094]

[0095] wherein, is the predicted value of the k+1th original data; is the predicted value of the first k+1 original accumulated data; is the predicted value of the first k original accumulated data.

[0096] The improved grey prediction model is corrected through the field measured load data, and the comparison is performed through the calculation of the residual error, and the residual error ε(k) is as follows:

[0097]

[0098] In the formula, x (0) (k) is the field measured load data, is the predicted load data, when ε(k) ≤ 0.1, the output requirement is reached, the values of λ and ζ are solved, if ε(k) > 0.1, the output requirement is not reached, the model is corrected according to the field measured value, that is, the values of λ and ζ are continuously adjusted.

[0099] The historical weather feature data cluster and the date feature data cluster with strong correlation and weak correlation in the meteorological observation station are input into the improved grey prediction model to obtain the predicted load value.

[0100] Step S6: According to the insulator conductance value and the amount of precipitation, the overhead ground wire electromagnetic induction voltage value and the static induction voltage value, the initial electric energy loss accurate prediction model is constructed.

[0101] S6 specifically includes:

[0102] Step S61: According to the electromagnetic coupling principle, a first relationship is constructed, and the electromagnetic induction voltage of the overhead ground wire of the power transmission line is calculated according to the first relationship; the first relationship is the relationship between the electromagnetic induction voltage of the overhead ground wire of the power transmission line and the first parameter of the power transmission line; the first parameter of the power transmission line includes the total number of towers of the power transmission line, the predicted load data, the line voltage, the power factor, the rotation factor, the distance between the overhead ground wire and the conductor in the base tower, and the electromagnetic induction voltage correction factor.

[0103] Step S62: According to the static induction principle, a second relationship is constructed, and the static induction voltage of the overhead ground wire of the power transmission line is calculated according to the second relationship; the second relationship is the relationship between the static induction voltage of the overhead ground wire of the power transmission line and the second parameter of the power transmission line; the second parameter of the power transmission line includes the charge coefficient of the conductor, the charge on the conductor, the charge coefficient on the overhead ground wire, the charge on the overhead ground wire in the base tower, and the static induction voltage correction factor.

[0104] Step S63: According to the electromagnetic induction voltage of the overhead ground wire of the power transmission line, the static induction voltage of the overhead ground wire of the power transmission line, and the first state of the first classification, a third relationship is determined, and the first electric energy loss of the overhead ground wire under the tower-by-tower grounding mode is calculated; the first classification is the overhead ground wire of the power transmission line under the tower-by-tower grounding mode; the first state is the non-idling state of the power transmission line; the third relationship is the relationship between the first electric energy loss of the overhead ground wire under the tower-by-tower grounding mode and the third parameter of the power transmission line; the third parameter of the power transmission line includes the electromagnetic induction voltage of the overhead ground wire, the static induction voltage of the overhead ground wire, the self-impedance of the overhead ground wire, the mutual impedance of the overhead ground wire, and the grounding resistance of the tower of the power transmission line.

[0105] Step S64: According to the second state of the first classification, a fourth relationship is determined, and the second electric energy loss of the overhead ground wire under the tower-by-tower grounding mode is calculated according to the fourth relationship; the second state is the idling state of the power transmission line; the fourth relationship is the relationship between the second electric energy loss of the overhead ground wire under the tower-by-tower grounding mode and the fourth parameter of the power transmission line; the fourth parameter of the power transmission line includes the charge on the overhead ground wire in the base tower, the angular frequency, and the grounding resistance of the base tower of the power transmission line.

[0106] Step S65: determining a fifth relationship according to the first state of the second classification, and calculating the conductance of the overhead ground wire insulator in the sectional insulation mode according to the fifth relationship; the second classification is the overhead ground wire of the transmission line in the sectional insulation mode; the fifth relationship is the relationship between the conductance of the overhead ground wire insulator in the sectional insulation mode and a fifth parameter of the transmission line; the fifth parameter includes the effective area of the insulator surface, the relative humidity of the weather, the rainfall intensity, and the surface conductivity variation coefficient.

[0107] Step S66: determining a sixth relationship according to the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, and the second state of the second classification, and calculating the electric energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode when the transmission line is in a non-idling state according to the sixth relationship; the sixth relationship is the relationship between the electric energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode when the transmission line is in a non-idling state and a sixth parameter; the sixth parameter includes the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, the conductance of the ground wire insulator affected by the precipitation, the self impedance of the overhead ground wire, and the sectional number.

[0108] Step S67: determining a seventh relationship according to the second state of the second classification, and calculating the electric energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode when the transmission line is in an idling state according to the seventh relationship; the seventh relationship is the relationship between the electric energy loss of the overhead ground wire affected by the precipitation in the sectional insulation mode when the transmission line is in an idling state and a seventh parameter; the seventh parameter includes the electrostatic induction voltage of the overhead ground wire of the transmission line and the conductance of the ground wire insulator affected by the precipitation.

[0109] Step S68: determining an eighth relationship according to the predicted load data and the predicted precipitation data meeting the second preset requirement, and selecting one from the candidate models as the initial electric energy loss accurate prediction model according to the eighth relationship; the candidate models include the fourth relationship, the fifth relationship, the sixth relationship, the seventh relationship, and 0.

[0110] In actual application, the construction process of the electric energy loss accurate prediction model specifically includes:

[0111] The overhead ground wire of the transmission line in the tower-by-tower grounding mode is classified into the first classification, and the overhead ground wire of the transmission line in the sectional insulation mode is classified into the second classification.

[0112] In normal operation, due to electromagnetic coupling and electrostatic induction effect, the overhead ground wire will generate electromagnetic induction voltage and electrostatic induction voltage, and under the action of the two, the overhead ground wire of the transmission line will generate electric energy loss.

[0113] According to the electromagnetic coupling principle, a first relationship is constructed, and according to the first relationship, the electromagnetic induction voltage of the overhead ground wire of the power transmission line is determined and a model is corrected through measurement data, the first relationship is represented by a first formula, and the first formula is:

[0114]

[0115] In the formula, E d is the induction voltage of the overhead ground wire, n is the total number of towers of the power transmission line, P is a predicted load obtained by calculation, U is a line voltage, cosφ is a power factor, θ is a rotation factor, d ia , d ib , and d ic are distances between the overhead ground wire and a, b, and c conductors in the i-th base tower, respectively; i is the i-th base tower of the power transmission line, j is a unit of imaginary number, and γ d is an electromagnetic induction voltage correction factor, which can be represented as l is a number of measurement data, and E d,measure is a measured electromagnetic induction voltage value.

[0116] According to the electrostatic induction principle, a second relationship is constructed, and according to the second relationship, the electrostatic induction voltage of the overhead ground wire of the power transmission line is determined and a model is corrected through measurement data, the second relationship is represented by a second formula, and the second formula is:

[0117]

[0118] In the formula, a a , a b , and a c are charge coefficients of a, b, and c three-phase conductors, Q ia , Q ib , and Q ic are charges on the respective lines of the a, b, and c three-phase conductors in the i-th base tower, a1 is a charge coefficient on the overhead ground wire, Q i1 is a charge on the overhead ground wire in the i-th base tower, and γ j is an electrostatic induction voltage correction factor, which can be represented as l is a number of measurement data, and E j,measure is a measured electrostatic induction voltage value.

[0119] The number l of measurement data of the measured electrostatic induction voltage value and the measured electromagnetic induction voltage value is the same, the electromagnetic induction voltage and the electrostatic induction voltage data are measured for each base tower of a line, and the measurement is performed simultaneously to ensure consistency of the line current.

[0120] A first state is divided from the first classification and the second classification, and the first state is a non-no-load condition of the power transmission line.

[0121] A second state is divided from both the first classification and the second classification, and the second state is a no-load condition of the power transmission line.

[0122] A third relationship is determined according to the first state under the first classification, and the electric energy loss of the overhead ground wire in the tower-by-tower grounding mode is calculated according to the third relationship, and the third relationship is represented by a third formula, and the third formula is:

[0123]

[0124] In the formula, P loss1 is the electric energy loss of the overhead ground wire in the tower-by-tower grounding mode when the power transmission line is not in a no-load condition, E i is the electromagnetic induction voltage of the overhead ground wire, E j is the electrostatic induction voltage of the overhead ground wire, Z self is the self-impedance of the overhead ground wire, Z mutual is the mutual impedance of the overhead ground wire, Z d is the grounding resistance of the tower of the power transmission line.

[0125] A fourth relationship is determined according to the second state under the first classification, and the electric energy loss of the overhead ground wire in the tower-by-tower grounding mode is calculated according to the fourth relationship, and the fourth relationship is represented by a fourth formula, and the fourth formula is:

[0126]

[0127] In the formula, P loss2 is the electric energy loss of the overhead ground wire in the tower-by-tower grounding mode when the power transmission line is in a no-load condition, Q i1 is the charge on the overhead ground wire in the i-th base tower, j is an imaginary number, w is an angular frequency, Z id is the grounding resistance of the i-th base tower of the power transmission line.

[0128] Rainfall has a significant impact on insulators, mainly because rainwater forms a water film on the surface of the insulator, thereby changing the conductance characteristics of the insulator, and further greatly affecting the electric energy loss of the overhead ground wire.

[0129] A fifth relationship is determined according to the first state under the second classification, and the conductance of the insulator of the overhead ground wire in the segmented insulation mode is calculated according to the fifth relationship, and the fifth relationship is represented by a fifth formula, and the fifth formula is:

[0130]

[0131] In the formula, G d is the conductance of the ground wire insulator affected by precipitation, A Sdenoted as the effective area of ​​the insulator surface, RH as the relative humidity in the predicted weather, RI as the rainfall intensity in the predicted weather, and k′ as the surface conductivity variation coefficient.

[0132] The sixth relationship can be determined based on the fifth relationship, and the overhead ground wire power loss under non-no-load conditions in segmented insulation mode can be calculated based on the sixth relationship. The sixth relationship is expressed by the sixth formula, which is:

[0133]

[0134] In the formula, P loss3 E represents the power loss of the overhead ground wire affected by precipitation under segmented insulation when the transmission line is not unloaded. i E is the electromagnetic induced voltage. j For electrostatic induced voltage, G d For the conductivity of the ground wire insulator affected by precipitation, Z self is the self-impedance of the overhead ground wire, and m′ is the number of segments.

[0135] The seventh relationship is determined based on the second state under the second category, and the overhead ground wire power loss under no-load conditions in the segmented insulation mode is calculated based on the seventh relationship. The seventh relationship is represented by the seventh formula, which is:

[0136] P loss4 =G d ×(E j ) 2 ;

[0137] In the formula, P loss4 When the transmission line is unloaded, the overhead ground wire energy loss due to precipitation under segmented insulation is E. j For electrostatic induced voltage, G d The conductivity of the ground wire insulator is affected by precipitation.

[0138] Based on the predicted load data and precipitation data in the predicted weather conditions of the dataset, an eighth relation is generated. The selection of the prediction model is then determined based on this eighth relation to achieve accurate predictions. The eighth relation is represented by an eighth formula, which is:

[0139]

[0140] Based on the predicted overhead ground wire power loss value and the actual overhead ground wire power loss value, a ninth relationship is generated, and the accuracy of the prediction model is determined based on the ninth relationship. The ninth relationship is represented by a ninth formula, which is:

[0141]

[0142] In the formula, Li actual loss, predicted loss, q is the number of loss data; E RMSE is the root mean square error.

[0143] According to the measured data and the calculated predicted power loss data, a tenth relationship is generated, and the prediction model is corrected according to the tenth relationship, and the tenth relationship is represented by a tenth formula, and the tenth formula is:

[0144] P correct = γ × P final ;

[0145] In the formula, P correct is the corrected predicted value, P final is the predicted value before correction, and γ is a correction factor, which can be represented as P measure is the measured value of power loss, and l is the number of measurement data.

[0146] Step S7: input the predicted weather feature data, the predicted load data and the target overhead ground wire parameter data of the transmission line into the power loss prediction model to obtain the power loss data of the target overhead ground wire of the transmission line; the power loss prediction model is obtained by training the initial power loss accurate prediction model using a power loss prediction training data set; the power loss prediction training data set includes the historical weather feature data, the historical load data, the target overhead ground wire parameter data of the transmission line and the corresponding historical power loss data of the target overhead ground wire of the transmission line; the target overhead ground wire parameter data of the transmission line includes the electromagnetic induction voltage value of the overhead ground wire, the electrostatic induction voltage value and the insulator conductance value.

[0147] In actual application, the power loss accurate prediction model is constructed according to the relationship between the insulator conductance value and the precipitation, and the relationship between the electromagnetic induction voltage value and the static induction voltage value and the load, the data set is input into the power loss accurate prediction model to obtain the power loss prediction data of the target overhead ground wire of the transmission line, and the prediction model is corrected by calculating the correction factor in combination with the actually measured power loss data.

[0148] Embodiment 2

[0149] A computer device, comprising: a memory, a processor to store a computer program on the memory and executable on the processor, and the processor executes the computer program to realize the power loss prediction method of the overhead ground wire of the transmission line in embodiment 1.

[0150] Embodiment 3

[0151] A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the power loss prediction method for overhead ground wire of power transmission line in embodiment 1.

[0152] Embodiment 4

[0153] A computer program product comprising a computer program which, when executed by a processor, implements the power loss prediction method for overhead ground wire of power transmission line in embodiment 1.

[0154] Embodiment 5

[0155] A computer device, which can be a database, can have an internal structure as shown in Figure 3 The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is used to store transactions to be processed. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor to implement the power loss prediction method for overhead ground wire of power transmission line in embodiment 1.

[0156] It should be noted that the object information (including but not limited to object device information, object personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the object or fully authorized by all parties, and the collection, use and processing of related data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0157] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, database or other medium used in the embodiments provided by the present application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided by the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided by the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0158] Any combination of the technical features of the above embodiments can be made. In order to make the description simple, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combination of the technical features does not exist, it should be considered as the scope of the present application.

[0159] The principles and implementation modes of the present application are described by using specific examples. The above description of the embodiments is only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation mode and application range will be changed. In conclusion, the content of the present application should not be understood as a limitation.

Claims

1. A method for predicting power loss in overhead ground wires of transmission lines, characterized in that, The power loss prediction method includes: The system acquires historical load data of the overhead ground wire of the target transmission line and historical weather characteristic data and corresponding date characteristic data of the target area where the overhead ground wire of the target transmission line is located; the historical weather characteristic data includes historical temperature data, historical precipitation data, historical humidity data, historical air pressure data, and historical wind speed data; the date characteristic data includes weekdays and non-weekdays. The historical load data and the historical weather characteristic data are preprocessed to obtain optimized historical load data and historical weather characteristic data; the preprocessing operation includes outlier removal and correction. Target weather feature data is selected from the optimized historical weather feature data; the correlation between the target weather feature data and the optimized historical load data meets the first preset requirement; Acquire predicted weather feature data that meets the second preset requirements, and apply a clustering algorithm to the predicted weather feature data that meets the second preset requirements and the corresponding date feature data to obtain a predicted weather feature data cluster and a date feature data cluster; the predicted weather feature data includes predicted temperature data, predicted precipitation data, predicted humidity data, predicted air pressure data and predicted wind speed data; The predicted weather feature data clusters and the date feature data clusters are input into the load forecasting model to obtain the predicted load data. The load forecasting model is obtained by training the improved grey forecasting model using the load forecasting training dataset. The load forecasting training dataset includes historical weather feature data cluster samples, corresponding date feature data cluster samples, and corresponding load data samples. The historical weather feature data cluster samples are obtained by applying a clustering algorithm to the target weather feature data. Based on the relationship between insulator conductivity and precipitation, and the relationship between electromagnetic induction voltage and electrostatic induction voltage of overhead ground wire and load, an accurate prediction model for initial power loss is constructed. The predicted weather feature data, the predicted load data, and the target transmission line overhead ground wire parameter data are input into the power loss prediction model to obtain the power loss data of the target transmission line overhead ground wire. The power loss prediction model is obtained by training the initial power loss accurate prediction model using a power loss prediction training dataset. The power loss prediction training dataset includes the historical weather feature data, the historical load data, the target transmission line overhead ground wire parameter data, and the corresponding historical power loss data of the target transmission line overhead ground wire. The target transmission line overhead ground wire parameter data includes the overhead ground wire electromagnetic induction voltage value, electrostatic induction voltage value, and insulator conductivity value.

2. The method for predicting power loss of overhead ground wires in transmission lines according to claim 1, characterized in that, The historical load data and historical weather characteristic data are preprocessed to obtain optimized historical load data and historical weather characteristic data, specifically including: Anomaly detection is performed on the historical load data and the historical weather characteristic data; Remove the outliers from the historical load data and the historical weather characteristic data; The historical load data and historical weather feature data after removing outliers are corrected by applying the mean algorithm to obtain optimized historical load data and historical weather feature data.

3. The method for predicting power loss of overhead ground wires in transmission lines according to claim 2, characterized in that, 3-sigma outlier detection is performed on the historical load data and the historical weather characteristic data.

4. The method for predicting power loss of overhead ground wires in transmission lines according to claim 1, characterized in that, Filtering target weather feature data from the historical weather feature data specifically includes: The correlation coefficient between the historical weather characteristic data and the historical load data was calculated using the Spearman correlation coefficient method. Based on the correlation coefficient, the correlation degree between the historical weather characteristic data and the historical load data is determined; the correlation degree includes strong correlation, weak correlation, and no correlation; when the correlation coefficient is greater than or equal to a first preset threshold, the correlation degree is strong correlation; when the correlation coefficient is less than the first preset threshold but greater than a second preset threshold, the correlation degree is weak correlation; when the correlation coefficient is less than or equal to the second preset threshold, the correlation degree is no correlation. Determine whether the degree of correlation is either strong or weak. When the correlation is strong or weak, the corresponding historical weather feature data is the target weather feature data.

5. The method for predicting power loss of overhead ground wires in transmission lines according to claim 4, characterized in that, Obtain predicted weather features that meet the second preset requirements, and apply a clustering algorithm to the date feature data and the predicted weather feature data that meet the second preset requirements to obtain a predicted weather feature data cluster and a date feature data cluster, specifically including: Acquire initial weather characteristic data of each meteorological monitoring station within the target area at each preset time within a preset time period; The initial weather feature data that are correlated with the historical load data to a degree that meets the second preset requirement are selected from the initial weather feature data to obtain the selected weather feature data; Calculate the average value of the filtered weather feature data to obtain the weather feature data of each preset time in the target area, and use the weather feature data of each preset time in the target area as the predicted weather feature data for the preset time period. The predicted weather feature data and the corresponding date feature data are processed using the k-means clustering algorithm to obtain a cluster of predicted weather feature data and a cluster of date feature data.

6. The method for predicting power loss of overhead ground wires in transmission lines according to claim 1, characterized in that, The construction process of the initial power loss accurate prediction model specifically includes: Based on the principle of electromagnetic coupling, a first relationship is constructed, and the electromagnetic induced voltage of the overhead ground wire of the transmission line is calculated based on the first relationship; the first relationship is the relationship between the electromagnetic induced voltage of the overhead ground wire of the transmission line and the first parameters of the transmission line; the first parameters of the transmission line include the total number of towers of the transmission line, predicted load data, line voltage, power factor, rotation factor, the spacing between the overhead ground wire and the conductor in the base tower, and the electromagnetic induced voltage correction factor. Based on the principle of electrostatic induction, a second relationship is constructed, and the electrostatic induced voltage of the overhead ground wire of the transmission line is calculated based on the second relationship; the second relationship is the relationship between the electrostatic induced voltage of the overhead ground wire of the transmission line and the second parameter of the transmission line; the second parameter of the transmission line includes the charge coefficient of the conductor, the charge on the conductor, the charge coefficient of the overhead ground wire, the charge on the overhead ground wire in the base tower, and the electrostatic induced voltage correction factor. Based on the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, and the first state of the first category, a third relationship is determined, and the first energy loss of the overhead ground wire under the tower-by-tower grounding method is calculated; the first category is the overhead ground wire of the transmission line under the tower-by-tower grounding method; the first state is the non-unloaded state of the transmission line; the third relationship is the relationship between the first energy loss of the overhead ground wire under the tower-by-tower grounding method and the third parameters of the transmission line; the third parameters of the transmission line include the electromagnetic induction voltage of the overhead ground wire, the electrostatic induction voltage of the overhead ground wire, the self-impedance of the overhead ground wire, the mutual impedance of the overhead ground wire, and the grounding resistance of the transmission line tower; Based on the second state of the first classification, a fourth relationship is determined, and the second power loss of the overhead ground wire under the tower-by-tower grounding method is calculated based on the fourth relationship; the second state is the no-load state of the transmission line; the fourth relationship is the relationship between the second power loss of the overhead ground wire under the tower-by-tower grounding method and the fourth parameter of the transmission line; the fourth parameter of the transmission line includes the charge, angular frequency and grounding resistance of the tower base of the transmission line on the overhead ground wire in the base tower; Based on the first state of the second category, a fifth relationship is determined, and the conductivity of the overhead ground wire insulator under the segmented insulation method is calculated based on the fifth relationship; the second category is the overhead ground wire of the transmission line with the segmented insulation form; the fifth relationship is the relationship between the conductivity of the overhead ground wire insulator under the segmented insulation method and the fifth parameter of the transmission line; the fifth parameter includes the effective area of ​​the insulator surface, the relative humidity of the weather, the rainfall intensity, and the surface conductivity variation coefficient; Based on the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, and the second state of the second category, a sixth relationship is determined, and the power loss of the overhead ground wire affected by precipitation under the segmented insulation method when the transmission line is not unloaded is calculated based on the sixth relationship; the sixth relationship is the relationship between the power loss of the overhead ground wire affected by precipitation under the segmented insulation method when the transmission line is not unloaded and the sixth parameter; the sixth parameter includes the electromagnetic induction voltage of the overhead ground wire of the transmission line, the electrostatic induction voltage of the overhead ground wire of the transmission line, the conductance of the ground wire insulator affected by precipitation, the self-impedance of the overhead ground wire, and the number of segments; Based on the second state of the second category, the seventh relationship is determined, and the overhead ground wire power loss affected by precipitation under the segmented insulation method when the transmission line is unloaded is calculated based on the seventh relationship; the seventh relationship is the relationship between the overhead ground wire power loss affected by precipitation under the segmented insulation method when the transmission line is unloaded and the seventh parameter; the seventh parameter includes the electrostatic induced voltage of the overhead ground wire of the transmission line and the conductance of the ground wire insulator affected by precipitation; Based on the predicted load data and the predicted precipitation data that meets the second preset requirements, an eighth relationship is determined, and one of the candidate models is selected as the initial accurate prediction model for power loss based on the eighth relationship; the candidate models include the fourth relationship, the fifth relationship, the sixth relationship, the seventh relationship, and 0.

7. The method for predicting power loss of overhead ground wires in transmission lines according to claim 6, characterized in that, The power loss prediction method also includes: The accuracy of the power loss prediction model is determined based on the power loss data of the target transmission line overhead ground wire output by the power loss prediction model and the actual power loss data of the target transmission line overhead ground wire.

8. The method for predicting power loss of overhead ground wires in transmission lines according to claim 1, characterized in that, The power loss prediction method also includes: Based on the predicted energy loss data and the corresponding measured energy loss data output by the initial energy loss accurate prediction model, the initial energy loss accurate prediction model is corrected.

9. A computer device, comprising: The memory and processor contain a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the power loss prediction method for overhead ground wires of transmission lines according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the computer program implements the method for predicting power loss of overhead ground wires of transmission lines as described in any one of claims 1-8.

Images