Power system load balancing control method based on data analysis

Abstract

The invention relates to the field of power grid load regulation, in particular to a power system load balancing control method based on data analysis. The method comprises the following steps: firstly, acquiring power supply voltages and power supply currents of different users on a power line in a to-be-tested power supply network at each moment, and load rates of a main transformer and the power line; obtaining a power supply insufficiency index of the target user according to the distribution of the power supply voltage and the power supply current of the target user and the local change of each abnormal voltage, clustering the users according to the power supply insufficiency index, obtaining a position distribution dispersion according to the position distribution of each user in each clustering cluster, and obtaining the position distribution dispersion according to the position distribution of each user in each clustering cluster; and in combination with the insufficient power supply index of the user, the position distribution dispersion of the cluster, and the load rates of the main transformer and the power line, carrying out balance control on the to-be-tested power supply network. According to the invention, the phenomena of overlarge load reduction degree of some users and concentrated load reduction can be avoided, and the load balancing control effect of the power system is improved.

Description

Technical Field

[0001] The present invention relates to the field of power grid load regulation, and specifically relates to a power system load balancing control method based on data analysis. Background Art

[0002] Power system load balancing control is an important part of the power management system. By various technologies and strategies, it adjusts the power load to meet the power supply and demand balance of the power grid. Power system load balancing control plays a crucial role in ensuring the reliability of power supply and improving the energy utilization rate of the power grid.

[0003] During the peak period of power demand, the power grid may face the risk of overload. In related technologies, measures are usually taken to cut the loads of some users in the power system, so that the load of the power system remains balanced. However, due to the large differences in the power supply situations of different users in the power system, it is easy to cut the loads of some users too much during the balancing control process, and it is easy to centrally cut the loads of users in the same area, resulting in poor effects on the power system load balancing control. Summary of the Invention

[0004] In order to solve the technical problems that in the balancing control process, it is easy to cut the loads of some users too much, and it is easy to centrally cut the loads of users in the same area, resulting in poor effects on the power system load balancing control, the purpose of the present invention is to provide a power system load balancing control method based on data analysis. The specific technical solutions adopted are as follows: The present invention proposes a power system load balancing control method based on data analysis. The method includes: Obtain the supply voltage and supply current of different users at each moment in each power line in the previous historical adjustment period of the power grid to be tested by the power company during the period to be adjusted, and obtain the load rate of each main transformer and the load rate of each power line in the historical adjustment period of the power grid to be tested; Take any power line as the target line and any user in the target line as the target user, and screen out the abnormal voltages in the supply voltages of the target user; perform autocorrelation analysis in the preset window of each abnormal voltage in the historical adjustment period of the target user to obtain the noise discrimination coefficient of each abnormal voltage; obtain the power supply shortage index of the target user according to the distribution of all the supply voltages of the target user, the distribution of the supply current, and the noise discrimination coefficient of each abnormal voltage; Cluster all users of the target line according to the power supply shortage index to obtain different clusters; obtain the position distribution dispersion of each cluster according to the position distribution of each user in each cluster; obtain the user adjustment priority of the target user according to the power supply shortage index of the target user and the position distribution dispersion of the cluster where the target user is located. Based on the user adjustment priorities of each user in each power line, the load factor of each main transformer in the power supply network to be measured, and the load factor of the power line, perform balanced control on the load of the power supply network to be measured during the adjustment period.

[0005] Further, the obtaining the noise discrimination coefficient of each abnormal voltage by performing autocorrelation analysis in a preset window of each abnormal voltage during the historical adjustment period of the target user includes: Take any abnormal voltage during the historical adjustment period of the target user as the target abnormal voltage, construct an autocorrelation function about the time delay variable according to the power supply voltages at each moment in the preset window of the target abnormal voltage, and obtain the autocorrelation coefficient of the target abnormal voltage under each time delay variable. Analyze the proportion of the number of autocorrelation coefficients of the target abnormal voltage that are less than the preset threshold in the total number of all autocorrelation coefficients to obtain the first noise discrimination factor of the target abnormal voltage. Analyze the overall level of all autocorrelation coefficients of the target abnormal voltage to obtain the overall autocorrelation coefficient of the target abnormal voltage. Obtain the noise discrimination coefficient of the target abnormal voltage according to the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage.

[0006] Further, the obtaining the noise discrimination coefficient of the target abnormal voltage according to the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage includes: Take the reciprocal of the sum of the overall autocorrelation coefficient of the target abnormal voltage and the preset first adjustment parameter to obtain the second noise discrimination factor of the target abnormal voltage. Synthesize the first noise discrimination factor and the second noise discrimination factor of the target abnormal voltage to obtain the noise discrimination coefficient of the target abnormal voltage.

[0007] Further, the obtaining the power supply shortage index of the target user according to the distribution of all the power supply voltages and the distribution of the power supply current of the target user and the noise discrimination coefficient of each abnormal voltage includes: Obtain the power supply instability coefficient of the target user according to the distribution of all the power supply voltages and the distribution of the power supply current of the target user. Input the noise discrimination coefficients of all abnormal voltages of the target user into the optimal threshold algorithm to obtain the optimal threshold, and use the abnormal voltages with noise discrimination coefficients less than the optimal threshold as the insufficient power supply voltages of the target user; Obtain the insufficient power supply index of the target user according to the quantity of the insufficient power supply voltages of the target user and the power supply instability coefficient.

[0008] Further, the obtaining of the power supply instability coefficient of the target user according to the distributions of all the power supply voltages and the power supply currents of the target user includes: Analyze the degree of dispersion of all the power supply voltages of the target user to obtain the power supply voltage instability coefficient of the target user; Analyze the degree of dispersion of all the power supply currents of the target user to obtain the power supply current instability coefficient of the target user; Integrate the power supply voltage instability coefficient and the power supply current instability coefficient of the target user to obtain the power supply instability coefficient of the target user.

[0009] Further, the obtaining of the insufficient power supply index of the target user according to the quantity of the insufficient power supply voltages of the target user and the power supply instability coefficient includes: Analyze the proportion of the quantity of the insufficient power supply voltages of the target user in the quantity of all the power supply voltages to obtain the insufficient power supply characteristic value of the target user; Integrate the insufficient power supply characteristic value and the power supply instability coefficient of the target user to obtain the insufficient power supply index of the target user.

[0010] Further, the obtaining of the distribution dispersion degree of each clustering cluster according to the position distributions of the users in each clustering cluster includes: Establish a two-dimensional coordinate system with the power company as the origin and determine the position coordinates of each user; The position coordinates include the abscissa value and the ordinate value. Take any clustering cluster as the target clustering cluster, analyze the degree of dispersion of the abscissa values of the position coordinates of all the users in the target clustering cluster to obtain the abscissa dispersion degree of the target clustering cluster, and analyze the degree of dispersion of the ordinate values of the position coordinates of all the users in the target clustering cluster to obtain the ordinate dispersion degree of the target clustering cluster; Integrate the abscissa dispersion degree and the ordinate dispersion degree of the target clustering cluster to obtain the first distribution dispersion degree of the target clustering cluster; In the target clustering cluster, divide the users corresponding to the position coordinates with the same abscissa value and the same ordinate value into the same category, and take the maximum value of the number of users in all categories as the distribution aggregation degree of the target clustering cluster; Obtain the location distribution dispersion degree of the target clustering cluster according to the first distribution dispersion degree and the distribution aggregation degree of the target clustering cluster.

[0011] Further, the obtaining the location distribution dispersion degree of the target clustering cluster according to the first distribution dispersion degree and the distribution aggregation degree of the target clustering cluster includes: Take the reciprocal of the sum of the distribution aggregation degree of the target clustering cluster and a preset second adjustment parameter to obtain the second distribution dispersion degree of the target clustering cluster; Integrate the first distribution dispersion degree and the second distribution dispersion degree of the target clustering cluster to obtain the location distribution dispersion degree of the target clustering cluster.

[0012] Further, the obtaining the user adjustment priority of the target user according to the power supply shortage index of the target user and the location distribution dispersion degree of the clustering cluster where the target user is located includes: Perform a negative correlation mapping on the power supply shortage index of the target user to obtain the necessity of load adjustment for the target user; Integrate and normalize the necessity of load adjustment of the target user and the location distribution dispersion degree of the clustering cluster where the target user is located to obtain the user adjustment priority of the target user.

[0013] Further, the performing balanced control on the load of the power supply network to be measured during the adjustment period includes: Integrate and normalize the user adjustment priorities of all users in the target line to obtain the line adjustment priority of the target line; In the power supply network to be measured, normalize the load rate of the main transformer connected to the target line to obtain the relative load degree of the main transformer connected to the target line; Integrate the load rate, the line adjustment priority, and the relative load degree of the target line to obtain the load distribution coefficient of the target line; Select a preset number of power lines with the largest load distribution coefficients from the power supply network to be measured as the lines to be distributed, and distribute the lines to be distributed to other connected power supply networks.

[0014] The present invention has the following beneficial effects: Considering that there are significant differences in the power supply situations of different users in the power system, which may lead to excessive load reduction for some users during the balancing control process and may also result in centralized load reduction for users in the same area, the present invention first obtains the supply voltage and supply current of different users at each moment in the previous historical adjustment period of the power company's power system during the period to be adjusted, and obtains the load rates of each main transformer and each power line in the measured power grid during the historical adjustment period. Since it is necessary to analyze the power supply shortage situation of the target user during the historical adjustment period, and power supply shortage may cause abnormal supply voltages for some users, the abnormal voltages of the target user can be obtained first. Considering that noise may also cause abnormal supply voltages, but the influence of noise will lead to irregular changes in local supply voltages, the possibility that the abnormal voltage of the target user is caused by power supply shortage can be reflected by the noise discrimination coefficient. Considering that when the power system has insufficient power supply to users, it will lead to a large difference between the supply current and supply voltage of users during the historical adjustment period, the power supply shortage situation of the power system for the target user can be reflected by the power supply shortage index. Then, users with similar power supply shortage indexes are grouped into the same clustering cluster. Considering that it is easy to conduct centralized load reduction for users in the same area during the power load balancing control process, the dispersion degree of the user location distribution in the same clustering cluster is reflected by the obtained location distribution dispersion degree. Then, based on the obtained user adjustment priorities, the load rates of each main transformer and each power line in the measured power grid, the load of the measured power grid during the period to be adjusted is balanced, avoiding excessive load reduction for some users and centralized load reduction for users in the same area during the balancing control process, and improving the effect of power system load balancing control. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0016] Figure 1 Flowchart of a method for power system load balancing control based on data analysis provided by an embodiment of the present invention; Figure 2 Flowchart of a method for obtaining the noise discrimination coefficient of each abnormal voltage provided by an embodiment of the present invention; Figure 3 Flowchart of a method for obtaining the power supply shortage index of the target user provided by an embodiment of the present invention; Figure 4Flowchart of the method for obtaining the dispersion of the position distribution of each clustering cluster provided by an embodiment of the present invention. Detailed implementation manners

[0017] In order to further elaborate on the technical means and effects adopted by the present invention to achieve the intended invention purpose, the following describes in detail a power system load balancing control method based on data analysis proposed according to the present invention, its specific implementation manners, structures, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0018] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs.

[0019] The following specifically describes the specific solution of a power system load balancing control method based on data analysis provided by the present invention in combination with the accompanying drawings.

[0020] Please refer to Figure 1 , which shows a flowchart of a power system load balancing control method based on data analysis provided by an embodiment of the present invention. The method includes: Step S1: Obtain the supply voltage and supply current of different users in each power line of the to-be-tested power supply network of the power company in the previous historical adjustment period of the to-be-adjusted period, and obtain the load rate of each main transformer and the load rate of each power line in the historical adjustment period of the to-be-tested power supply network.

[0021] Power system load balancing control is an important part of the power management system. By various technologies and strategies, it adjusts the power load to meet the power supply and demand balance of the power grid. Power system load balancing control plays a crucial role in ensuring the reliability of power supply and improving the energy utilization rate of the power grid. During the peak period of power demand, the power grid may face the risk of overload. In related technologies, usually measures are taken to cut the load of some users in the power system, thereby reducing the load of the power system and keeping the load of the power system balanced. However, due to the large differences in the power supply situations of different users in the power system, it is easy to cut the load of some users too much during the balancing control process, and it is easy to centrally cut the load of users in the same area, thus affecting the normal power consumption of users and resulting in poor control effect of the power system load balancing. Therefore, the embodiments of the present invention propose a power system load balancing control method based on data analysis to solve this problem.

[0022] Since multiple power supply networks erected by power companies in the power system will provide power services to multiple users simultaneously, in the embodiments of the present invention, power monitoring devices in the power system are first used to collect the power supply voltage and power supply current of different users at each moment in the previous historical adjustment period of the power supply network to be measured during the current adjustment period to be adjusted. The time interval for collecting data is set to 1 second. In other embodiments of the present invention, it can also be set by the implementer according to the specific implementation scenario and is not limited herein. The length of the adjustment period to be adjusted and the historical adjustment period are equal, and the length range is generally , in hours. In one embodiment of the present invention, the lengths of both are set to 2 hours. In other embodiments of the present invention, it can also be set by the implementer according to the specific implementation scenario and is not limited herein.

[0023] Under normal circumstances, the voltage provided by the power system for each user does not remain constant but fluctuates within a certain range. Therefore, it is also necessary to obtain the nominal voltage lower limit value of each power line in the power supply network to be measured. When the voltage provided by the power system for a certain user is insufficient, the power supply voltage of this user will be less than the nominal voltage lower limit value of the power line where it is located, which is convenient for subsequent analysis of the insufficient power supply situation of the power system for users based on the power supply voltage of the user less than the nominal voltage lower limit value, avoiding excessive load reduction for some users and reducing the effect of the final power load balancing control. Regarding the description of the nominal voltage lower limit value: for example, the nominal voltage of the power line used by ordinary residential users is 220V, and its nominal voltage lower limit value is generally 90% of the nominal voltage.

[0024] At the same time, it is also necessary to obtain the load rate of each main transformer and the load rate of each power line in the power supply network to be measured during the historical adjustment period. The load rates of the main transformer and the power line are usually recorded by relevant devices in the power system, which is convenient for subsequent analysis of the load conditions of the main transformer and the power line in the power supply network to be measured, so as to balance the load of the power line with high load.

[0025] Step S2: Select any power line as the target line and any user in the target line as the target user, and screen out the abnormal voltages in the power supply voltage of the target user; perform autocorrelation analysis in the preset window of each abnormal voltage in the historical adjustment period of the target user to obtain the noise discrimination coefficient of each abnormal voltage; obtain the power supply shortage index of the target user according to the distribution of all the power supply voltages and power supply currents of the target user and the noise discrimination coefficient of each abnormal voltage.

[0026] In the subsequent steps of the embodiments of the present invention, it is necessary to analyze the power supply shortage situation of each user in each power line during the historical adjustment period, so as to achieve load shedding for different users according to different priorities during the period to be adjusted, and avoid excessive load shedding for some users. Therefore, for the convenience of more clear analysis in the subsequent steps, any power line can be first taken as the target line, and any user in the target line can be taken as the target user. The insufficient power supply will cause the partial power supply voltage of the target user to be less than the lower limit value of the nominal voltage of the target line. Therefore, the power supply voltage of the target user that is less than the lower limit value of the nominal voltage of the target line can be used as the abnormal voltage of the target user. At the same time, considering that the existence of noise will also cause the partial power supply voltage of the target user to be less than the lower limit value of the nominal voltage, that is to say, the abnormal voltage may be caused by the insufficient power supply of the power system to the target user, or may be caused by the influence of noise during the data acquisition process. Therefore, in order to more accurately analyze the situation of the power system's insufficient power supply to the target user during the historical adjustment period, it is necessary to further distinguish the abnormal voltages generated by the two different factors.

[0027] Due to the randomness of noise, its influence on the power supply voltage of the target user will cause irregular changes in the local power supply voltage during the historical adjustment period. When the power system has insufficient power supply to the target user, it will cause the local power supply voltage to change regularly or periodically to a certain extent during the historical adjustment period. Therefore, autocorrelation analysis can be performed in the preset window of each abnormal voltage in the historical adjustment period of the target user. The obtained noise discrimination coefficient reflects the possibility of each abnormal voltage being generated by the influence of noise, which is convenient for subsequent distinguishing the abnormal voltage generated by noise and the abnormal voltage generated by insufficient power supply based on the noise discrimination coefficient, and improving the accuracy of analyzing the situation of insufficient power supply to the target user. The length of the preset window is set to 60, that is, the preset window includes the abnormal voltage itself and the 59 power supply voltages closest to the abnormal voltage. The length of the preset window can also be set by the implementer according to the specific implementation scenario, and is not limited here.

[0028] Preferably, in an embodiment of the present invention, the method for obtaining the noise discrimination coefficient of each abnormal voltage specifically includes: Please refer to Figure 2 , which shows the flowchart of the method for obtaining the noise discrimination coefficient of each abnormal voltage provided by an embodiment of the present invention.

[0029] Step S201: Take any abnormal voltage in the historical adjustment period of the target user as the target abnormal voltage, construct an autocorrelation function about the time-delay variable according to the power supply voltages at each moment in the preset window of the target abnormal voltage, and obtain the autocorrelation coefficient of the target abnormal voltage at each time-delay variable.

[0030] First, it is necessary to analyze a certain abnormal voltage. Therefore, any abnormal voltage in the historical adjustment period of the target user can be used as the target abnormal voltage. As can be seen from the above analysis, the influence of noise on the power supply voltage of the target user will cause irregular changes in the local power supply voltage during the historical adjustment period. When the power system supplies insufficient power to the target user, it will cause the change of the local power supply voltage during the historical adjustment period to have a certain degree of regularity. The autocorrelation function can reflect the regularity of the time series change by calculating and analyzing the correlation between data at different time delay moments in the time series. Therefore, an autocorrelation function about the time delay variable can be constructed according to the power supply voltage at each moment in the preset window of the target abnormal voltage, and the autocorrelation coefficient of the target abnormal voltage at each time delay variable can be obtained through the constructed autocorrelation function. The smaller the autocorrelation coefficient at each time delay variable, the less regular the change of the power supply voltage in the preset window of the target abnormal voltage at that time delay variable, and the more likely the target abnormal voltage is caused by noise influence. Subsequently, based on the autocorrelation coefficients of the target abnormal voltage at different time delay variables as the data basis, the noise discrimination coefficient of the target abnormal voltage can be accurately calculated.

[0031] The expression of the autocorrelation function of the target abnormal voltage in the preset window can be specifically, for example: Among them, represents the autocorrelation function of the target abnormal voltage in the preset window, and its value range is , and the independent variable is the time delay variable ; represents the power supply voltage at the th moment in the preset window of the target abnormal voltage; represents the power supply voltage at the th moment in the preset window of the target abnormal voltage; represents the average value of all power supply voltages in the preset window of the target abnormal voltage; represents the number of all moments in the preset window of the target abnormal voltage, that is, the number of power supply voltages in the preset window of the target abnormal voltage; represents the variance of the power supply voltages at all moments in the preset window of the target abnormal voltage.

[0032] Substitute the values of the time delay variable one by one into the above constructed autocorrelation function, and the autocorrelation coefficient of the target abnormal voltage at each time delay variable can be calculated.

[0033] Step S202: Analyze the proportion of the number of autocorrelation coefficients of the target abnormal voltage that are less than the preset threshold in the number of all autocorrelation coefficients to obtain the first noise discrimination factor of the target abnormal voltage.

[0034] The closer the autocorrelation coefficient under each time delay variable is to 0, it indicates that under this time delay variable, the change of the power supply voltage in the preset window of the target abnormal voltage is more irregular. Therefore, among all the autocorrelation coefficients of the target abnormal voltage, the more autocorrelation coefficients close to 0, it indicates that the target abnormal voltage is more likely to be generated by noise interference. Therefore, the proportion of the number of autocorrelation coefficients of the target abnormal voltage less than the preset threshold in the number of all autocorrelation coefficients can be analyzed, and the first noise discrimination factor obtained is used to reflect the possibility that the target abnormal voltage is generated by noise interference, providing a data basis for calculating the noise discrimination coefficient of the target abnormal voltage in the subsequent process.

[0035] In the embodiment of the present invention, the value of the preset threshold should be as small as possible, and its value range is generally , in an embodiment of the present invention, the preset threshold is set to 0.05. The preset threshold can also be set by the implementer according to the specific implementation scenario, and is not limited herein.

[0036] In an embodiment of the present invention, the number of autocorrelation coefficients of the target abnormal voltage less than the preset threshold can be used as the numerator, and the number of all autocorrelation coefficients of the target abnormal voltage can be used as the denominator, and the ratio is used as the first noise discrimination factor of the target abnormal voltage, so as to realize the analysis of its proportion. Among them, the number of all autocorrelation coefficients of the target abnormal voltage is .

[0037] Step S203: Analyze the overall level of all autocorrelation coefficients of the target abnormal voltage to obtain the overall autocorrelation coefficient of the target abnormal voltage.

[0038] The smaller the autocorrelation coefficient under each time delay variable, it indicates that under this time delay variable, the change of the power supply voltage in the preset window of the target abnormal voltage is more irregular, and the target abnormal voltage is more likely to be generated by noise interference. Therefore, the overall level of all autocorrelation coefficients of the target abnormal voltage can be analyzed to obtain the overall autocorrelation coefficient of the target abnormal voltage. The smaller the overall autocorrelation coefficient, it indicates that the overall level of all autocorrelation coefficients of the target abnormal voltage is smaller, and further indicates that the target abnormal voltage is more likely to be generated by noise interference, providing a data basis for calculating the noise discrimination coefficient of the target abnormal voltage in the subsequent process.

[0039] In the embodiment of the present invention, the average value or median of all autocorrelation coefficients of the target abnormal voltage can be used as the overall autocorrelation coefficient of the target abnormal voltage, so as to realize the analysis of the overall level of all autocorrelation coefficients of the target abnormal voltage, and is not limited herein.

[0040] Step S204: Obtain the noise discrimination coefficient of the target abnormal voltage according to the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage.

[0041] The larger the first noise discrimination factor of the target abnormal voltage and the smaller its overall autocorrelation coefficient, the more likely it is that the target abnormal voltage is caused by noise. Therefore, based on the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage, the noise discrimination coefficient of the target abnormal voltage can be obtained. The larger the noise discrimination coefficient, the more likely it is that the target abnormal factor is caused by noise and the less likely it is caused by insufficient power supply. Subsequently, based on the noise discrimination coefficient, the abnormal voltage caused by noise and the abnormal voltage caused by insufficient power supply can be accurately distinguished, improving the accuracy of analyzing the insufficient power supply situation of the target user.

[0042] Preferably, in an embodiment of the present invention, the method for obtaining the noise discrimination coefficient of the target abnormal voltage specifically includes: Taking the reciprocal of the sum of the overall autocorrelation coefficient of the target abnormal voltage and a preset first adjustment parameter to obtain the second noise discrimination factor of the target abnormal voltage. The larger the second noise discrimination factor, the more likely it is that the target abnormal voltage is caused by noise.

[0043] Combining the first noise discrimination factor and the second noise discrimination factor of the target abnormal voltage to obtain the noise discrimination coefficient of the target abnormal voltage. The larger the noise discrimination coefficient, the more likely it is that the target abnormal factor is caused by noise and the less likely it is caused by insufficient power supply.

[0044] In the embodiment of the present invention, the first noise discrimination factor and the second noise discrimination factor of the target abnormal voltage can be multiplied or added to achieve the combination of the two, which is not limited herein.

[0045] The expression of the noise discrimination coefficient of the target abnormal voltage can be specifically, for example: Wherein, represents the noise discrimination coefficient of the target abnormal voltage of the target user; represents the number of autocorrelation coefficients of the target abnormal voltage that are less than the preset threshold; represents the number of all moments in the preset window of the target abnormal voltage, represents the number of all autocorrelation coefficients of the target abnormal voltage; represents the overall autocorrelation coefficient of the target abnormal voltage; represents the first noise discrimination factor of the target abnormal voltage; represents the second noise discrimination factor of the target abnormal voltage; represents the preset first adjustment parameter to prevent the denominator from being 0, The value range of In an embodiment of the present invention, Set to 0.01. In other embodiments of the present invention, it can also be set by the implementer according to the specific implementation scenario, and is not limited herein.

[0046] It should be noted that in other embodiments of the present invention, the positive and negative correlation relationship can also be characterized by other basic mathematical operations, which will not be elaborated herein.

[0047] By the same method as above, the noise discrimination coefficient of each abnormal voltage of the target user can be obtained. The smaller the noise discrimination coefficient of the abnormal voltage at a certain moment, the more likely it is that the abnormal voltage of the target user at that moment is caused by insufficient power supply of the power system. At the same time, considering that when the power system supplies insufficient power to the user, it will cause a large difference in the supply current at each moment in the historical adjustment period of the user, and a large difference in the supply voltage at each moment. Therefore, the distribution of all supply voltages and supply currents of the target user and the noise discrimination coefficient of each abnormal voltage can be analyzed, and the power supply shortage index obtained is used to reflect the situation of the power system supplying insufficient power to the target user, providing a data basis for subsequent analysis of the priority of different user load curtailment.

[0048] Preferably, in an embodiment of the present invention, the method for obtaining the power supply shortage index of the target user specifically includes: Please refer to Figure 3 , which shows the flowchart of the method for obtaining the power supply shortage index of the target user provided by an embodiment of the present invention.

[0049] Step S211: Obtain the power supply instability coefficient of the target user according to the distribution of all supply voltages and supply currents of the target user.

[0050] When the power system supplies insufficient power to the target user, it will cause a large difference in the supply current at each moment in the historical adjustment period of the target user, and a large difference in the supply voltage at each moment. Therefore, the distribution of all supply voltages and supply currents of the target user can be analyzed first, and the power supply instability coefficient obtained is used to reflect the degree of insufficient or unstable power supply of the power system to the target user in the historical adjustment period, providing a data basis for subsequent calculation and analysis of the power supply shortage index of the target user.

[0051] Preferably, in an embodiment of the present invention, the method for obtaining the power supply instability coefficient of the target user specifically includes: When the power system supplies insufficient power to the target user, it will cause a large difference in the supply voltage at each moment in the historical adjustment period of the target user. Therefore, the dispersion degree of all supply voltages of the target user can be analyzed to obtain the power supply voltage instability coefficient. The larger the power supply voltage instability coefficient, the greater the dispersion degree of all supply voltages of the target user, and the more unstable the power supply situation of the power system to the target user in the historical adjustment period.

[0052] In the embodiment of the present invention, statistical quantities such as the standard deviation, variance, or range of all the supply voltages of the target user can be used as the supply voltage instability coefficient of the target user to analyze the degree of dispersion of the supply voltage, which is not limited herein.

[0053] Similarly, when the power system supplies insufficient power to the target user, it will cause a large difference in the supply current at each moment during the historical adjustment period of the target user. Therefore, the degree of dispersion of all the supply currents of the target user can be analyzed to obtain the supply current instability coefficient of the target user. The larger the supply current instability coefficient, the greater the degree of dispersion of all the supply currents of the target user, and the more unstable the power supply situation of the power system to the target user during the historical adjustment period.

[0054] To improve the calculation accuracy, in the embodiment of the present invention, the same statistical quantity as that used to calculate the supply voltage instability coefficient of the target user is used to calculate the supply current instability coefficient of the target user. For example, if the standard deviation of all the supply voltages of the target user is used as the supply voltage instability coefficient, then the standard deviation of all the supply currents of the target user is also used as the supply current instability coefficient.

[0055] The larger the supply voltage instability coefficient and the supply current instability coefficient, the more unstable the power supply situation of the power system to the target user during the historical adjustment period. Therefore, the supply voltage instability coefficient and the supply current instability coefficient of the target user can be combined to obtain the supply instability coefficient of the target user.

[0056] In the embodiment of the present invention, the supply voltage instability coefficient and the supply current instability coefficient of the target user can be multiplied or added to achieve the combination of the two, which is not limited herein.

[0057] Step S212: Input the noise discrimination coefficients of all the abnormal voltages of the target user into the optimal threshold algorithm to obtain the optimal threshold, and use the abnormal voltages with noise discrimination coefficients less than the optimal threshold as the insufficient supply voltages of the target user.

[0058] During the historical adjustment period of the target user, the abnormal voltage with a smaller noise discrimination coefficient is more likely to be generated due to insufficient power supply from the power system to the target user. To improve the discrimination accuracy, the noise discrimination coefficients of all abnormal voltages of the target user can be input into the optimal threshold algorithm to obtain the optimal threshold, and the abnormal voltages with noise discrimination coefficients less than the optimal threshold are regarded as the insufficient power supply voltages of the target user. The larger the number of insufficient power supply voltages, the more frequent the insufficient power supply from the power system to the target user during the historical adjustment period. Subsequently, the insufficient power supply index of the target user can be accurately calculated by combining the number of insufficient power supply voltages and the power supply instability coefficient, realizing the analysis of the insufficient power supply situation of the power system to the target user, and providing a data basis for subsequent analysis of the priority of load reduction for different users.

[0059] In the embodiment of the present invention, the optimal threshold algorithm can select existing algorithms such as the maximum inter-class variance algorithm or the maximum entropy method, etc., which is not limited herein.

[0060] Step S213: Obtain the insufficient power supply index of the target user according to the number of insufficient power supply voltages and the power supply instability coefficient of the target user.

[0061] The larger the number of insufficient power supply voltages of the target user, the more frequent the insufficient power supply from the power system to the target user during the historical adjustment period, and the larger the power supply instability coefficient of the target user, the greater the degree of insufficient or unstable power supply from the power system to the target user during the historical adjustment period. Therefore, the insufficient power supply index of the target user can be obtained according to the number of insufficient power supply voltages and the power supply instability coefficient of the target user. The insufficient power supply index reflects the insufficient power supply situation of the power system to the target user during the historical adjustment period. The larger the insufficient power supply index, the greater the degree of insufficient power supply from the power system to the target user during the historical adjustment period.

[0062] Preferably, in an embodiment of the present invention, the method for obtaining the insufficient power supply index of the target user further includes: Analyze the proportion of the number of insufficient power supply voltages of the target user in the number of all power supply voltages to obtain the insufficient power supply characteristic value of the target user. The larger the insufficient power supply characteristic value, the more frequent the insufficient power supply from the power system to the target user during the historical adjustment period, and thus the greater the degree of insufficient power supply from the power system to the target user during the historical adjustment period.

[0063] In an embodiment of the present invention, the number of insufficient power supply voltages of the target user can be used as the numerator, the number of all power supply voltages of the target user can be used as the denominator, and the ratio is used as the insufficient power supply characteristic value of the target user, so as to realize the analysis of its quantity proportion.

[0064] Integrate the insufficient power supply characteristic value and the power supply instability coefficient of the target user to obtain the insufficient power supply index of the target user.

[0065] In an embodiment of the present invention, the power supply shortage eigenvalue of the target user and the power supply instability coefficient can be multiplied or added to achieve the integration of the two.

[0066] The expression of the power supply shortage index of the target user can be specifically, for example: Wherein, represents the power supply shortage index of the target user; represents the number of power supply shortage voltages of the target user; represents the number of all power supply voltages of the target user; represents the power supply instability coefficient of the target user; represents the power supply shortage eigenvalue of the target user.

[0067] By the above same method, the power supply shortage index of each user can be obtained. Subsequently, based on the power supply shortage index, the priority of power load reduction for different users can be evaluated to avoid excessive power load reduction for some users, thus resulting in a poor problem of the load balancing control effect of the power system.

[0068] Step S3: Cluster all users of the target line according to the power supply shortage index to obtain different clusters; according to the location distribution of each user in each cluster, obtain the location distribution dispersion degree of each cluster; according to the power supply shortage index of the target user and the location distribution dispersion degree of the cluster where the target user is located, obtain the user adjustment priority of the target user.

[0069] Since the obtained power supply shortage index can reflect the power supply shortage situation of each user in the historical regulation period of the power system, for some users with similar power supply shortage indexes, it indicates that the power supply shortage situations of these users by the power system are similar. Therefore, in order to analyze the users with similar power supply shortage situations in the target line simultaneously, all users of the target line can be clustered according to the power supply shortage index to obtain different clusters. The power supply shortage situations of each user in the same cluster are relatively similar, and then the location distribution situations of each user in the same cluster can be analyzed.

[0070] Preferably, in an embodiment of the present invention, the existing k-means algorithm can be used to cluster users with similar power supply shortage indexes. In other embodiments of the present invention, other clustering algorithms such as the DBSCAN algorithm can also be used, which is not limited herein.

[0071] Considering that in the process of power load balancing control, it is easy to centrally cut the loads of users in the same area, thereby reducing the effect of power load balancing control. Therefore, it is also necessary to obtain the dispersion of the position distribution of each cluster according to the position distribution of each user in each cluster, and use the dispersion of the position distribution to reflect the dispersion degree of the position distribution of each user in each cluster. Subsequently, different priorities can be set for different users in the cluster based on the dispersion of the position distribution to avoid the occurrence of centralized load reduction and improve the effect of power load balancing control.

[0072] Preferably, in an embodiment of the present invention, the method for obtaining the dispersion of the position distribution of each cluster specifically includes: Please refer to Figure 4 , which shows the flowchart of the method for obtaining the dispersion of the position distribution of each cluster provided by an embodiment of the present invention.

[0073] Step S301: Establish a two-dimensional coordinate system with the power company as the origin, determine the position coordinates of each user, where the position coordinates include the abscissa value and the ordinate value. Take any one cluster as the target cluster, analyze the dispersion degree of the abscissa values of the position coordinates of all users in the target cluster to obtain the abscissa dispersion of the target cluster, and analyze the dispersion degree of the ordinate values of the position coordinates of all users in the target cluster to obtain the ordinate dispersion of the target cluster.

[0074] In an embodiment of the present invention, the horizontal axis of the two-dimensional coordinate system is the east-west direction, and the vertical axis is the north-south direction. In other embodiments of the present invention, other directions can also be used as the horizontal axis and the vertical axis of the two-dimensional coordinate system, as long as the horizontal axis and the vertical axis are perpendicular to each other, which is not limited herein.

[0075] To analyze a specific cluster more clearly, any one cluster can be taken as the target cluster. Since the position coordinates of each user include the abscissa value and the ordinate value, the dispersion degree of the abscissa values of the position coordinates of all users in the target cluster can be analyzed. The obtained abscissa dispersion is used to reflect the dispersion degree of the abscissa values of each user in the target cluster, and the dispersion degree of the ordinate values of the position coordinates of all users in the target cluster is analyzed. The obtained ordinate dispersion is used to reflect the dispersion degree of the ordinate values of each user in the target cluster, providing a data basis for subsequent analysis of the dispersion degree of the position distribution of each user in the target cluster.

[0076] In the embodiments of the present invention, statistics such as standard deviation, variance, or range can be used to analyze the dispersion degree of the abscissa values or ordinate values of each user in the target cluster. And to improve the subsequent calculation accuracy, the same statistic needs to be used to analyze the dispersion degree of both.

[0077] Step S302: Synthesize the abscissa dispersion and ordinate dispersion of the target clustering cluster to obtain the first distribution dispersion of the target clustering cluster.

[0078] The greater the abscissa dispersion, it indicates that the abscissa values of the users in the target clustering cluster are more dispersed. The greater the ordinate dispersion, it indicates that the abscissa values of the users in the target clustering cluster are more dispersed. Therefore, the abscissa dispersion and ordinate dispersion of the target clustering cluster can be synthesized to obtain the first distribution dispersion of the target clustering cluster, and the first distribution dispersion is used to reflect the dispersion degree of the positions of the users in the target clustering cluster, providing a data basis for calculating the position distribution dispersion of the target clustering cluster in the subsequent process.

[0079] In the embodiments of the present invention, the abscissa dispersion and ordinate dispersion of the target clustering cluster can be added or multiplied to achieve the synthesis of the two, and no limitation is made here.

[0080] Step S303: In the target clustering cluster, divide the users corresponding to the position coordinates with the same abscissa value and the same ordinate value into the same category, and take the maximum value of the number of users in all categories as the distribution aggregation degree of the target clustering cluster.

[0081] Considering that the position coordinates of some users in the target clustering cluster may be the same, for example, users living in the same building. Therefore, in the target clustering cluster, the users corresponding to the position coordinates with the same abscissa value and the same ordinate value can also be divided into the same category, and take the maximum value of the number of users in all categories as the distribution aggregation degree of the target clustering cluster. The greater the distribution aggregation degree, it indicates that the number of users with the same position coordinates in the target clustering cluster is more, and further indicates that the dispersion degree of the positions of the users in the target clustering cluster is smaller. Subsequently, the position distribution dispersion of the target clustering cluster can be calculated and analyzed by combining the distribution aggregation degree of the target clustering cluster and the first distribution dispersion obtained above.

[0082] Step S304: Obtain the position distribution dispersion of the target clustering cluster according to the first distribution dispersion and the distribution aggregation degree of the target clustering cluster.

[0083] The greater the first distribution dispersion and the smaller the distribution aggregation degree, it indicates that the dispersion degree of the positions of the users in the target clustering cluster is greater. Therefore, the position distribution dispersion of the target clustering cluster can be obtained according to the first distribution dispersion and the distribution aggregation degree of the target clustering cluster, and the position distribution dispersion accurately reflects the dispersion degree of the positions of the users in the target clustering cluster.

[0084] Preferably, in an embodiment of the present invention, the method for obtaining the position distribution dispersion of the target clustering cluster specifically includes: Take the reciprocal of the sum of the distribution concentration degree of the target clustering cluster and the preset second adjustment parameter to obtain the second distribution dispersion degree of the target clustering cluster. Synthesize the first distribution dispersion degree and the second distribution dispersion degree of the target clustering cluster to obtain the position distribution dispersion degree of the target clustering cluster. In the embodiments of the present invention, the first distribution dispersion degree and the second distribution dispersion degree of the target clustering cluster can be multiplied or added to achieve the synthesis of the two, which is not limited herein.

[0085] As an example, in an embodiment of the present invention, the expression of the position distribution dispersion degree of the target clustering cluster can be specifically, for example: Wherein, represents the position distribution dispersion degree of the target clustering cluster; represents the abscissa dispersion degree of the target clustering cluster; represents the ordinate dispersion degree of the target clustering cluster; represents the distribution concentration degree of the target clustering cluster, that is, the maximum value of the number of users in all categories in the target clustering cluster; represents the first distribution dispersion degree of the target clustering cluster; represents the second distribution dispersion degree of the target clustering cluster; represents the preset second adjustment parameter to prevent the denominator from being zero, The value range of In an embodiment of the present invention, is set to 0.01. In other embodiments of the present invention, it can also be set by the implementer according to the specific implementation scenario, which is not limited herein.

[0086] It should be noted that in other embodiments of the present invention, other basic mathematical operations can also be used to represent the positive and negative correlation relationship, which will not be elaborated herein.

[0087] Through the above same method, the position distribution dispersion degree of each clustering cluster can be obtained. In the process of taking measures to reduce the load of some users in the power system so as to keep the load of the power system balanced, in order to avoid excessive reduction of the power load of some users and the occurrence of concentrated load reduction phenomenon, the user adjustment priority of the target user can be obtained according to the power supply shortage index of the target user and the position distribution dispersion degree of the clustering cluster where the target user is located. The smaller the user adjustment priority, the smaller the priority of reducing the power load of the target user relative to other users. Subsequently, different priorities of power load reduction can be performed on various users based on the user adjustment priority, so as to improve the effect of load balance control of the power system.

[0088] Preferably, in an embodiment of the present invention, the method for obtaining the user adjustment priority of the target user specifically includes: The larger the power supply shortage index of the target user, the greater the degree of power supply shortage of the power system to the target user during the historical adjustment period. Therefore, during the period to be adjusted, it is necessary to reduce the priority of power load reduction for the target user to avoid excessive power load reduction for the target user, thus affecting the normal power consumption of the target user. Therefore, a negative correlation mapping can be performed on the power supply shortage index of the target user to obtain the necessity of load adjustment for the target user. The greater the necessity of load adjustment, the greater the necessity of power load reduction for the target user, that is, it is necessary to give priority to power load reduction for the target user.

[0089] The greater the position distribution dispersion of the clustering cluster where the target user is located, the more discrete the position distribution between the target user and other users in this clustering cluster. Therefore, it is necessary to give priority to power load reduction for the target user to avoid the occurrence of centralized power load reduction for users. Therefore, the necessity of load adjustment for the target user and the position distribution dispersion of the clustering cluster where the target user is located can be combined and then normalized to obtain the user adjustment priority of the target user. The greater the user adjustment priority, the more necessary it is to give priority to power load reduction for the target user. Subsequently, based on the user adjustment priority, different priorities of power load reduction can be performed on each user to reduce the impact on the normal power consumption of users and improve the effect of load balancing control of the power system.

[0090] In the embodiments of the present invention, the necessity of load adjustment for the target user and the position distribution dispersion of the clustering cluster where the target user is located can be multiplied or added to achieve the combination of the two, and no limitation is made here.

[0091] In an embodiment of the present invention, the normalization process can be, for example, the maximum-minimum normalization process. And, the normalization in subsequent steps can all adopt the maximum-minimum normalization process. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of values, and this will not be elaborated here.

[0092] The expression of the user adjustment priority of the target user can be, for example: Among them, represents the user adjustment priority of the target user; represents the power supply shortage index of the target user; represents the position distribution dispersion of the clustering cluster where the target user is located; represents the necessity of load adjustment for the target user; represents the normalization function; represents a preset third adjustment parameter to prevent the denominator from being zero, The value range of In an embodiment of the present invention, Set to 0.01, and in other embodiments of the present invention, it can also be set by the implementer according to the specific implementation scenario, which is not limited herein.

[0093] By the same method as above, the user adjustment priority of each user in the target line can be obtained.

[0094] Step S4: Based on the user adjustment priority of each user in each power line, the load rate of each main transformer in the power supply network to be measured, and the load rate of the power line, perform load balancing control on the load of the power supply network to be measured during the period to be adjusted.

[0095] During the process of load balancing control of the power system, the user adjustment priority can reflect the priority of power load reduction for each user in the target line, and the load rate of each main transformer and the load rate of the power line in the power supply network to be measured can reflect the real load situation of the main transformer and the power line during operation. Therefore, based on the user adjustment priority of each user in each power line, the load rate of each main transformer in the power supply network to be measured, and the load rate of the power line, load balancing control can be performed on the load of the power supply network to be measured during the period to be adjusted, so as to avoid excessive load reduction for some users during the balancing control process and the occurrence of concentrated load reduction for users, reduce the impact on the normal power consumption of each user in the power system, and can effectively reduce the load of the power supply network to be measured and improve the effect of load balancing control of the power system.

[0096] Preferably, in an embodiment of the present invention, the method for performing load balancing control on the load of the power supply network to be measured during the period to be adjusted specifically includes: After comprehensively combining the user adjustment priorities of all users in the target line and performing normalization processing, the line adjustment priority of the target line is obtained. The larger the line adjustment priority, the more it indicates that the load of the target line needs to be adjusted preferentially.

[0097] In the embodiment of the present invention, the comprehensive combination of the user adjustment priorities of all users in the target line can be achieved by calculating the cumulative value or product value of the user adjustment priorities of all users in the target line, which is not limited herein.

[0098] In the power supply network to be measured, the load rate of the main transformer connected to the target line is normalized to obtain the relative load degree of the main transformer connected to the target line. The larger the relative load degree, the greater the load of the main transformer connected to the target line relative to other main transformers.

[0099] In one embodiment of the present invention, the load rate of the main transformer connected to the target line can be used as the numerator, the cumulative value of the load rates of all the main transformers in the power grid to be measured can be used as the denominator, and the ratio can be used as the relative load level of the main transformer connected to the target line, so as to realize the normalization processing of the load rate of the main transformer connected to the target line.

[0100] Meanwhile, the larger the load rate of the target line is, the greater the load on the target line is. Therefore, the load rate of the target line, the line adjustment priority, and the relative load level can be integrated to obtain the load distribution coefficient of the target line. The larger the load distribution coefficient is, the more necessary it is to balance the load of the target line.

[0101] In the embodiment of the present invention, the sum value or product value of the load rate of the target line, the line adjustment priority, and the relative load level of the main transformer connected to the target line can be used as the load distribution coefficient of the target line to realize the integration of the three, which is not limited herein.

[0102] Furthermore, a preset number of power lines with the largest load distribution coefficients can be selected from the power grid to be measured as the lines to be distributed, and the lines to be distributed can be distributed to other connected power grids. While realizing the load balance of the power grid to be measured, it avoids excessive load reduction for some users during the balance control process and the occurrence of the phenomenon of centralized load reduction for users, improving the effect of load balance control of the power system. The preset number is set to 10, and the specific value of the preset number can also be set by the implementer according to the specific implementation scenario, which is not limited herein.

[0103] In summary, in the embodiment of the present invention, first, the supply voltage and supply current of different users at each moment in each power line in the previous historical adjustment period of the power grid to be measured of the power company in the period to be adjusted are obtained, as well as the nominal voltage lower limit value of each power line, and the load factor of each main transformer and the load factor of each power line in the historical adjustment period of the power grid to be measured are obtained; a two-dimensional coordinate system is established with the power company as the origin to determine the position coordinates of each user; any one power line is used as the target line, any one user in the target line is used as the target user, and the supply voltage lower than the nominal voltage lower limit value of the target line in the target user is used as the abnormal voltage of the target user; autocorrelation analysis is performed in the preset window of each abnormal voltage in the historical adjustment period of the target user to obtain the noise discrimination coefficient of each abnormal voltage; according to the distribution of all the supply voltages and supply currents of the target user and the noise discrimination coefficient of each abnormal voltage, the power supply shortage index of the target user is obtained; clustering is performed on all the users of the target line according to the power supply shortage index to obtain different clustering clusters; according to the distribution of the position coordinates of the users in each clustering cluster, the position distribution dispersion of each clustering cluster is obtained; according to the power supply shortage index of the target user and the position distribution dispersion of the clustering cluster where the target user is located, the user adjustment priority of the target user is obtained; based on the user adjustment priorities of the users in each power line, the load factor of each main transformer and the load factor of the power line in the power grid to be measured, the load of the power grid to be measured in the period to be adjusted is balanced and controlled.

[0104] It should be noted that the above sequence of the embodiments of the present invention is only for description and does not represent the advantages or disadvantages of the embodiments. The processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0105] Each embodiment in this specification is described in a progressive manner. The same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A power system load balancing control method based on data analysis, characterized in that, The method includes: Obtaining the supply voltage and supply current of different users at each moment in each power line during the previous historical adjustment period of the power supply network to be measured by the power company in the to-be-adjusted period, and obtaining the load rate of each main transformer and the load rate of each power line in the power supply network to be measured during the historical adjustment period; Taking any one power line as the target line and any one user in the target line as the target user, screening out the abnormal voltages in the supply voltages of the target user; performing autocorrelation analysis in the preset window of each abnormal voltage in the historical adjustment period of the target user to obtain the noise discrimination coefficient of each abnormal voltage; obtaining the power supply shortage index of the target user according to the distribution of all the supply voltages of the target user, the distribution of the supply current, and the noise discrimination coefficient of each abnormal voltage; Clustering all users of the target line according to the power supply shortage index to obtain different clustering clusters; obtaining the position distribution dispersion degree of each clustering cluster according to the position distribution of each user in each clustering cluster; obtaining the user adjustment priority of the target user according to the power supply shortage index of the target user and the position distribution dispersion degree of the clustering cluster where the target user is located; Based on the user adjustment priorities of each user in each power line, the load rate of each main transformer and the load rate of the power line in the power supply network to be measured, perform balanced control on the load of the power supply network to be measured in the to-be-adjusted period.

2. The power system load balancing control method based on data analysis according to claim 1, wherein The performing autocorrelation analysis in the preset window of each abnormal voltage in the historical adjustment period of the target user to obtain the noise discrimination coefficient of each abnormal voltage includes: Taking any one abnormal voltage in the historical adjustment period of the target user as the target abnormal voltage, constructing an autocorrelation function about the time delay variable according to the supply voltages at each moment in the preset window of the target abnormal voltage, and obtaining the autocorrelation coefficient of the target abnormal voltage at each time delay variable; Analyzing the proportion of the number of autocorrelation coefficients of the target abnormal voltage that are less than the preset threshold in the total number of all autocorrelation coefficients to obtain the first noise discrimination factor of the target abnormal voltage; Analyzing the overall level of all autocorrelation coefficients of the target abnormal voltage to obtain the overall autocorrelation coefficient of the target abnormal voltage; Obtaining the noise discrimination coefficient of the target abnormal voltage according to the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage.

3. A method for controlling the load balance of a power system based on data analysis according to claim 2, characterized in that, The obtaining the noise discrimination coefficient of the target abnormal voltage according to the first noise discrimination factor and the overall autocorrelation coefficient of the target abnormal voltage includes: Taking the reciprocal of the sum value of the overall autocorrelation coefficient of the target abnormal voltage and the preset first adjustment parameter to obtain the second noise discrimination factor of the target abnormal voltage; Integrating the first noise discrimination factor and the second noise discrimination factor of the target abnormal voltage to obtain the noise discrimination coefficient of the target abnormal voltage.

4. A method for controlling the load balance of a power system based on data analysis according to claim 1, characterized in that The obtaining the power supply shortage index of the target user according to the distribution of all the supply voltages of the target user, the distribution of the supply current, and the noise discrimination coefficient of each abnormal voltage includes: Obtain the power supply instability coefficient of the target user according to the distribution of the power supply voltage and the distribution of the power supply current of the target user; Input the noise discrimination coefficient of all abnormal voltages of the target user into the optimal threshold algorithm to obtain the optimal threshold, and regard the abnormal voltage whose noise discrimination coefficient is less than the optimal threshold as the insufficient power supply voltage of the target user; Obtain the insufficient power supply index of the target user according to the number of the insufficient power supply voltages of the target user and the power supply instability coefficient; 5. A method for power system load balancing control based on data analysis according to claim 4, characterized in that The obtaining of the power supply instability coefficient of the target user according to the distribution of the power supply voltage and the distribution of the power supply current of the target user includes: Analyze the dispersion degree of all power supply voltages of the target user to obtain the power supply voltage instability coefficient of the target user; Analyze the dispersion degree of all power supply currents of the target user to obtain the power supply current instability coefficient of the target user; Integrate the power supply voltage instability coefficient and the power supply current instability coefficient of the target user to obtain the power supply instability coefficient of the target user; 6. The method for controlling the load balance of a power system based on data analysis according to claim 4, wherein The obtaining of the insufficient power supply index of the target user according to the number of the insufficient power supply voltages of the target user and the power supply instability coefficient includes: Analyze the proportion of the number of the insufficient power supply voltages of the target user in the number of all power supply voltages to obtain the insufficient power supply characteristic value of the target user; Integrate the insufficient power supply characteristic value and the power supply instability coefficient of the target user to obtain the insufficient power supply index of the target user; 7. A method for load balancing control of a power system based on data analysis according to claim 1, characterized in that The obtaining of the position distribution dispersion degree of each clustering cluster according to the position distribution of each user in each clustering cluster includes: Establish a two-dimensional coordinate system with the power company as the origin to determine the position coordinates of each user; The position coordinates include the abscissa value and the ordinate value. Take any clustering cluster as the target clustering cluster, analyze the dispersion degree of the abscissa values of the position coordinates of all users in the target clustering cluster to obtain the abscissa dispersion degree of the target clustering cluster, and analyze the dispersion degree of the ordinate values of the position coordinates of all users in the target clustering cluster to obtain the ordinate dispersion degree of the target clustering cluster; Integrate the abscissa dispersion degree and the ordinate dispersion degree of the target clustering cluster to obtain the first distribution dispersion degree of the target clustering cluster; In the target clustering cluster, divide the users corresponding to the position coordinates with the same abscissa value and the same ordinate value into the same category, and take the maximum value of the number of users in all categories as the distribution aggregation degree of the target clustering cluster; Obtain the position distribution dispersion degree of the target clustering cluster according to the first distribution dispersion degree and the distribution aggregation degree of the target clustering cluster; 8. A method for load balancing control of a power system based on data analysis according to claim 7, characterized in that, The obtaining of the position distribution dispersion degree of the target clustering cluster according to the first distribution dispersion degree and the distribution aggregation degree of the target clustering cluster includes: Take the reciprocal of the sum value of the distribution aggregation degree of the target clustering cluster and the preset second adjustment parameter to obtain the second distribution dispersion degree of the target clustering cluster; Integrate the first distribution dispersion degree and the second distribution dispersion degree of the target clustering cluster to obtain the position distribution dispersion degree of the target clustering cluster; 9. A method for controlling the load balance of a power system based on data analysis according to claim 1, characterized in that, Obtaining the user adjustment priority of the target user according to the power supply shortage index of the target user and the location distribution dispersion degree of the cluster where the target user is located includes: Performing a negative correlation mapping on the power supply shortage index of the target user to obtain the necessity of load adjustment of the target user; After comprehensively considering the necessity of load adjustment of the target user and the location distribution dispersion degree of the cluster where the target user is located and performing normalization processing, obtaining the user adjustment priority of the target user.

10. A method for load balancing control of a power system based on data analysis according to claim 1, characterized in that, The balanced control of the load of the power grid to be measured during the adjustment period includes: After comprehensively considering the user adjustment priorities of all users in the target line and performing normalization processing, obtaining the line adjustment priority of the target line; In the power grid to be measured, performing normalization processing on the load rate of the main transformer connected to the target line to obtain the relative load degree of the main transformer connected to the target line; Comprehensively considering the load rate, the line adjustment priority, and the relative load degree of the target line to obtain the load distribution coefficient of the target line; Selecting a preset number of power lines with the largest load distribution coefficients from the power grid to be measured as the lines to be allocated, and allocating the lines to be allocated to other connected power grids.

Images