Low-voltage power grid fault diagnosis method and system based on improved whale optimization algorithm

Abstract

The application provides a low-voltage power grid fault diagnosis method and system based on an improved whale optimization algorithm, and the topology structure of a low-voltage power grid is obtained according to low-voltage power grid station area data; based on the obtained topology structure, the impedance of each branch of the low-voltage power grid is calculated by using an improved whale optimization algorithm based on Levy flight and simulated annealing; according to the impedance of each branch, the power loss of each branch is calculated by using a root mean square current method; and according to the power loss of each branch, the low-voltage power grid fault diagnosis result is obtained; the Levy flight strategy and the simulated annealing strategy are applied to the optimization process of the whale optimization algorithm, so that the whale optimization algorithm can jump out of the local optimum and the global search ability is enhanced, and the impedance calculation method based on the improved whale optimization algorithm based on Levy flight and simulated annealing is proposed by combining the line flow calculation, accurate calculation of the line loss of the station area branch is realized, and more accurate monitoring and fault diagnosis of the low-voltage power grid are realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electric power engineering, in particular to a low-voltage power grid fault diagnosis method and system based on an improved whale optimization algorithm. BACKGROUND

[0002] The statements in this section merely provide background technology related to the present application and do not necessarily constitute prior art.

[0003] In a low-voltage distribution line, the network structure is complex, and the working state of the distribution line is difficult to monitor. At the same time, as the use time of the distribution line increases, aging occurs, user equipment fails, electricity theft occurs, etc., which will all cause losses to the power grid. Therefore, real-time monitoring of the line state is of great significance to ensure the safe and stable operation of the power grid.

[0004] The inventor found that the traditional line flow calculation method is tedious, low in accuracy and not suitable for large-scale transformer area calculation, while the whale optimization algorithm can improve the efficiency and accuracy of the flow calculation. However, the traditional whale algorithm has the limitations of low precision, difficulty in achieving global optimization, and easy to fall into premature convergence. SUMMARY

[0005] In order to solve the problems of the prior art, the present application provides a low-voltage power grid fault diagnosis method and system based on an improved whale optimization algorithm. The Levy flight strategy and simulated annealing strategy are applied to the whale optimization algorithm optimization process, so that it can jump out of the local optimum and enhance the global search ability. The impedance calculation method based on Levy flight and simulated annealing improved whale optimization algorithm is proposed by combining it with the line flow calculation, which realizes the accurate calculation of the transformer area branch line loss, and further realizes the more accurate monitoring and fault diagnosis of the low-voltage power grid.

[0006] In order to achieve the above purpose, the present application adopts the following technical solutions:

[0007] In a first aspect, the present application provides a low-voltage power grid fault diagnosis method based on an improved whale optimization algorithm.

[0008] A low-voltage power grid fault diagnosis method based on an improved whale optimization algorithm, comprising the following processes:

[0009] Obtain low-voltage power grid transformer area data;

[0010] Obtain the topology structure of the low-voltage power grid according to the low-voltage power grid transformer area data;

[0011] Based on the obtained topology structure, calculate the impedance of each branch of the low-voltage power grid using the whale optimization algorithm improved based on Levy flight and simulated annealing;

[0012] According to the impedance of each branch, calculate the power loss of each branch using the root mean square current method;

[0013] According to the branch power loss, a low-voltage power grid fault diagnosis result is obtained.

[0014] As a further limitation of the first aspect of the application, the low-voltage power grid substation data at least includes the voltage, current, active power and reactive power of each branch of the substation.

[0015] As a further limitation of the first aspect of the application, after obtaining the low-voltage power grid substation data, the low-voltage power grid substation data is preprocessed.

[0016] The preprocessing at least includes using a Bayesian linear regression model to fill in the low-voltage power grid substation data.

[0017] As a further limitation of the first aspect of the application, the mean of the standard deviation in all time dimensions is used as the objective function, and the whale optimization algorithm improved based on Levy flight and simulated annealing is used to calculate the impedance of each branch of the low-voltage power grid.

[0018] Wherein, the standard deviation is used to represent the difference between the calculated value and the true value of the transformer secondary side voltage derived from the user data, and the mean function of the standard deviation in all time dimensions is obtained according to the standard deviation corresponding to all users at the same time. The function is used as the objective function.

[0019] As a further limitation of the first aspect of the application, the whale optimization algorithm improved based on Levy flight and simulated annealing is used to calculate the impedance of each branch of the low-voltage power grid, including:

[0020] According to the determined objective function, a random population of independent variables is generated;

[0021] The value of the objective function is calculated, and the optimal function value and the corresponding position are recorded, and the values of the first matrix vector coefficient, the second matrix vector coefficient, the weight factor and the whale change behavior probability are updated;

[0022] When the whale change behavior probability is less than the first set value and the first matrix vector coefficient is less than the second set value, the surrounding hunting update position is performed;

[0023] When the whale change behavior probability is less than the first set value and the first matrix vector coefficient is greater than or equal to the second set value, the random search update position is performed;

[0024] When the whale change behavior probability is greater than or equal to the first set value and the first matrix vector coefficient is greater than or equal to the second set value, the spiral update position is performed;

[0025] A new population is generated, the function value of the updated objective function is calculated, and the corresponding position is recorded;

[0026] If the function value of the updated target function is better than the optimal value, the new population whale position replaces the original whale position;

[0027] If the function value of the updated target function is worse than the optimal value, the acceptance of the poor solution probability is calculated to determine whether to accept the new population whale position;

[0028] The temperature reduction operation is performed, Levy flight is performed, and the self position is updated;

[0029] It is judged whether the maximum iteration number is reached, if not, the iteration calculation is returned, and if the maximum iteration number is reached, the next step is performed;

[0030] The global optimal value and the position of the corresponding independent variable are output, and then the matrix of the branch impedance in the transformer area is obtained.

[0031] As a further limitation of the first aspect of the application, the first matrix coefficient A is A=2ar1-a, and the second matrix coefficient C is C=2r2, wherein a=2-2n / N max , n is the iteration number, r1 and r2 are random numbers between 0 and 1, a is a coefficient that decreases linearly with the increase of the iteration number, and N max represents the maximum iteration number.

[0032] As a further limitation of the first aspect of the application, according to the branch impedance, the root mean square current method is used to calculate the branch power loss, including:

[0033] According to the matrix of the branch impedance in the transformer area and the voltage, active power and reactive power of each branch, the power loss matrix of each branch is obtained.

[0034] Secondly, the application provides a low-voltage power grid fault diagnosis system based on the improved whale optimization algorithm.

[0035] A low-voltage power grid fault diagnosis system based on the improved whale optimization algorithm, comprising:

[0036] The data acquisition module is configured to acquire low-voltage power grid area data;

[0037] The topology generation module is configured to obtain the topology structure of the low-voltage power grid according to the low-voltage power grid area data;

[0038] The impedance calculation module is configured to calculate the branch impedance of the low-voltage power grid based on the obtained topology structure by using the whale optimization algorithm improved based on Levy flight and simulated annealing;

[0039] The power loss calculation module is configured to calculate the branch power loss by using the root mean square current method according to the branch impedance;

[0040] The fault diagnosis module is configured to obtain a low-voltage power grid fault diagnosis result according to the power losses of the branches.

[0041] In a third aspect, the present application provides a computer readable storage medium having a program stored thereon, which, when executed by a processor, implements the steps of the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm according to the first aspect of the present application.

[0042] In a fourth aspect, the present application provides an electronic device comprising a memory, a processor, and a program stored on the memory and executable on the processor, wherein the processor implements the steps of the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm according to the first aspect of the present application when executing the program.

[0043] Compared with the prior art, the present application has the following advantages:

[0044] 1. The Levy flight strategy and the simulated annealing strategy are applied to the whale optimization algorithm in the present application, which expands the diversity of the algorithm population and can accept inferior solutions with a certain probability, thereby improving the global search ability of the algorithm and jumping out of the local optimum to achieve the global optimum.

[0045] 2. The whale optimization algorithm based on Levy flight and simulated annealing is applied to the impedance calculation step of the line load flow calculation method in the present application, which can efficiently and accurately calculate the impedance values of each branch in the substation, and enhances the ability to calculate large-scale substation impedance. Combined with the root mean square current method, the power loss of each branch can be quickly and accurately analyzed.

[0046] The advantages of the additional aspects of the present application will be partially given in the following description, partially become obvious from the following description, or be understood through the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0047] The accompanying drawings, which form a part of the specification, are included to provide a further understanding of the application and are incorporated herein by reference. The illustrations are shown for the purpose of explaining the present application and should not be construed as an improper limitation of the present application.

[0048] Figure 1 A flowchart of the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm provided for the first embodiment of the present application is shown in the figure.

[0049] Figure 2 The null abnormal data preprocessing effect provided for the first embodiment of the present application is shown in the figure.

[0050] Figure 3 The impedance model of the distribution network line provided for the first embodiment of the present application is shown in the figure.

[0051] Figure 4A line impedance model between two points of a power distribution network line is provided for the embodiment 1 of the present application.

[0052] Figure 5 A voltage U&1 is provided for the embodiment 1 of the present application. a A vector relationship diagram of the voltage U&1.

[0053] Figure 6 A flowchart of the improved whale optimization algorithm is provided for the embodiment 1 of the present application. DETAILED DESCRIPTION

[0054] The present application will be further described below in conjunction with the accompanying drawings and embodiments.

[0055] It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the present application. Unless otherwise indicated, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application pertains.

[0056] It should be noted that the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular form is intended to include the plural form unless the context clearly indicates otherwise, and furthermore, it should be understood that when the terms "comprise" and / or "include" are used in the specification, there is a presence of the features, steps, operations, devices, components, and / or combinations thereof.

[0057] The embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0058] Embodiment 1

[0059] The embodiment 1 of the present application provides a low-voltage power grid fault diagnosis method based on an improved whale optimization algorithm. The Levy flight strategy and the simulated annealing strategy are applied to the whale optimization algorithm optimization process, so that it can jump out of the local optimum and enhance the global search ability. The impedance calculation method based on the Levy flight and simulated annealing improved whale optimization algorithm is proposed by combining it with the line flow calculation, and the accurate calculation of the line loss of the transformer branch is realized.

[0060] Specifically, as shown in the figure, the following processes are included: Figure 1

[0061] S1: Collecting transformer area data and performing data preprocessing

[0062] The power grid system collects a large amount of user power consumption data in real time, including the voltage, current, active power and reactive power of each branch of the transformer area, etc. Due to the complex and changeable collection environment, equipment failure and other problems, the voltage, current and power data often appear missing and affect the subsequent analysis and calculation. ​

[0063] In this invention, a Bayesian linear regression (BLR) model is used for data imputation during data preprocessing. To verify the accuracy of this method in imputing data, a missing data imputation experiment was conducted, and the experimental results are as follows: Figure 2 As shown in the image, this method can effectively restore and fill in missing data.

[0064] S2: Analysis of the topological structure of the transformer area

[0065] The power supply network topology is obtained based on user data provided by the power grid, and an impedance model of the distribution network lines is established, such as... Figure 3 As shown in the figure, a represents the transformer substation, b and c represent line nodes, and Z... ij The numbers represent the impedances of each branch in the topology, and the end numbers 1, 2, 3... represent the user numbers.

[0066] S3: Impedance Calculation Method Based on Levy Flight and Simulated Annealing Improved Whale Optimization Algorithm (L-SA-IWOA)

[0067] The resistance between any two points in a power grid line, such as Figure 3 China Z d1 It can be used Figure 4 The model shown describes the impedance Z. d1 It can be expressed by equation (1):

[0068] Z d1 =R1+jX1 (1)

[0069] Where R1 and X1 represent the resistance and reactance of the line, respectively.

[0070] At any time t, the line impedance Z is given by Ohm's law for the line as shown in equation (2):

[0071]

[0072] In the formula: These represent the voltages at node d and node 1, respectively. Z represents the current flowing from node d to node 1. d1 This indicates the impedance of that section of the line.

[0073] Voltage in equation (2) and The vector relationship can be described as Figure 5 As shown in the figure, The direction is the reference axis, and δ1 represents and The included angle, express and The included angle will Decomposed into co-directional and perpendicular to directional perpendicular to The expressions of the longitudinal and transverse components of voltage drop are as follows:

[0074]

[0075]

[0076] Wherein, the intermediate variables used in calculating the longitudinal and transverse components of voltage drop are obtained by decomposing into R1I1 and jX1I1, the direction of R1I1 is the same as that of and the direction of jX1I1 is perpendicular to that of .

[0077] In the model of Figure 3 , the power expression of user node 1 is as follows:

[0078]

[0079]

[0080] Wherein, U1 represents user voltage, I1 represents user current, P1 is active power and Q1 is reactive power.

[0081] According to the vector relationship of Figure 5 , the following formula can be obtained:

[0082]

[0083] By combining formulas (1) to (7), the following formula can be obtained:

[0084]

[0085] Wherein, U d represents upstream node voltage, U1, I1, P1 and Q1 represent respective known parameters of downstream node.

[0086] Figure 3 The line between node a and node d in the model can also be expressed as the voltage U a of node a, and the equation about the resistance and reactance between node a and node d; wherein, U d of node d can be calculated by the data of all users downstream of node d, U d can be obtained by formula (8), P d and Q d can be calculated by the data of all users downstream of node d, and U d can be obtained by formula (8). dThe sum of the power of all users downstream of point d and the loss power is equal to d, and the specific calculation method is as follows:

[0087]

[0088]

[0089] Wherein, m represents the total number of user branches downstream of point d, P j , Q j represent the active and reactive power of downstream users, ΔP j , ΔP j represent the loss power corresponding to each downstream user branch.

[0090] Using the collected existing data and formula (9) and (10), the voltage, active power and reactive power of each branch node in the model shown in formula (11) can be obtained Figure 3 Using the existing data to perform a line flow calculation from bottom to top, from each user end, a calculated value U ai of the secondary side voltage of a transformer can be derived from bottom to top, and all U ai obtained can be represented by a voltage matrix [U a1 , U a2 , …, U an ] T , wherein i = 1, 2, …, n represents the number of downstream users, and n is the total number of users.

[0091] The time interval of the collected user data is 15 minutes, and through the flow calculation at different times t, the U ait at the corresponding time can be obtained, so the U ait of the entire area at different times can be represented as:

[0092]

[0093] Wherein, n represents the total number of users at time t, and T represents the time series length for flow calculation of the i-th user branch.

[0094] There is a difference between the result obtained through the flow calculation and the true value, in order to measure the error between each calculated value U ait and the actual measured value, the standard deviation (Standard Deviation, SD) is used to represent the calculation error, and in the voltage matrix, the SD of all user calculation values at the same time t is as follows:

[0095]

[0096] Wherein, i represents the number of user ends, n represents the total number of users, and (U at ) measuredU(t) represents the real measurement value of the transformer secondary side voltage at this moment.

[0097] The error δ of SD is calculated as follows under the whole time sequence:

[0098]

[0099] Where, T represents the length of time sequence, (U at ) SD represents the standard deviation of all user calculation values at any t moment.

[0100] The value of δ reflects the accuracy of the impedance calculation result, the closer the value of δ to 0, the more accurate the impedance calculation result, so formula (14) is taken as the objective function of L-SA-IWOA algorithm:

[0101]

[0102] Where, (U at ) measured represents the measured value of U at , (U at ) SD represents the value of standard deviation, δ represents the mean value of standard deviation in all time dimensions.

[0103] The improved whale optimization algorithm based on Levy flight and simulated annealing (L-SA-IWOA) obtains the optimal solution of the objective function through three stages of whale surrounding prey, spiral updating and random search. The simulated annealing and Levy flight strategies are introduced on the basis of the classical whale optimization algorithm to further improve the global search ability of the whale optimization algorithm.

[0104] S3.1: Surrounding prey stage

[0105] Let x = [R1, R2, L, R n , Z1, Z2, L, Z n ] be the independent variable, and the resistance and reactance of each branch are regarded as the independent variable, i.e. the position of the whale in the search range, and the position of the whale closest to the prey in the search range is X(n), and the corresponding objective function solution is the optimal solution F best This stage continuously approaches the prey through the sharing of the position information of each whale and the updating of the optimal position of the whale. The position updating of this stage is as follows:

[0106] x(n+1) = X(n) - A * D + λ (15)

[0107] D1 = |C * X(n) - x(n)| (16)

[0108] λ = F * (X(n) - x(n)) (17)

[0109] where n represents the iteration number; x(n) represents the position of the whale at the nth iteration; X(n) represents the optimal solution position at the nth iteration; D1 represents the difference between x(n) and X(n); λ represents a differential mutation perturbation factor, which enables the whale individual to jump out of the local optimum, improving the optimization accuracy and efficiency of the algorithm; F is a mutation degree factor; A and C are matrix vector coefficients as follows:

[0110]

[0111] C = 2r2 (19)

[0112] where r1 and r2 are random numbers between 0 and 1, a is a coefficient that decreases linearly with the increase of the iteration number, N max represents the maximum iteration number, and A essentially represents a random number in [-a, a].

[0113] S3.2: Spiral update phase

[0114] After determining the prey, the optimal individual in the population also performs spiral update towards the optimal position, and the spiral update method is as follows:

[0115] x(n+1) = X(n) + ω * D2 * e bl *cos(2πl) (20)

[0116] D2 = |X(n) - x(n)| (21)

[0117]

[0118] where D2 is the distance between the current position and the optimal position; b is a spiral shape coefficient; l is a random number in [-1, 1]; ω represents a weight factor as shown in (22), and k is an adjustment coefficient.

[0119] The whale hunting process will spiral update its position around the prey and continuously shrink the surrounding circle to move towards it, and use a bubble net to attack the prey during hunting. In order to balance the behavior mechanisms of surrounding shrinkage and spiral updating position, a critical value of 0.5 is applied when |A| < 1 to determine whether to surround hunting or spiral update position, and the specific method is as follows:

[0120]

[0121] where p is the probability of the whale changing behavior, and is a random number in [0, 1].

[0122] S3.3: Random search phase

[0123] The surrounding prey stage and spiral update stage are local search of the whale approaching the target. To improve the global search ability of the algorithm, when |A|≥1, the whale's target is far away, and the whale position is randomly selected from the search range for search. The specific method is as follows:

[0124] x(n+1)=x rand -ω*A*D3 (24)

[0125] D3=|C*x rand -x(n)| (25)

[0126] Wherein, x is the current whale position; x rand is a randomly selected individual position; ω represents the weight factor as shown in equation (22); D3 represents the difference between the random position and the current position.

[0127] S3.4: Simulated annealing

[0128] According to the Metropolis criterion, the probability of the particle tending to be stable at T temperature is as shown in equation (26):

[0129]

[0130] Where ΔE is the change of internal energy, and K is the Boltzmann constant. In order to make the algorithm accept the inferior solution with a certain probability and jump out of the local optimum to a certain extent, the acceptance probability P inferior is as shown in equation (27):

[0131]

[0132] Where F(x(n)) is the function value of the nth iteration, and t is the temperature of the nth iteration. The simulated annealing is introduced into the whale algorithm to generate a new random population x', calculate the target function value F'(x), and record the corresponding position. When the whale individuals of the new population are worse than the original population, it is determined whether to accept the new population whale position according to equation (27). The temperature reduction process of simulated annealing is as shown in equation (28):

[0133] t n =0.99*t n-1 (28)

[0134] Where t n represents the temperature of the nth annealing, and the initial temperature t0 is defined as the difference between the maximum and minimum values of the target function value of the initial population of the algorithm t0=F max (x)-F min (x).

[0135] S3.5: Levy flight updates position

[0136] A big defect of the classic whale optimization algorithm is difficult to achieve global optimization, easy to fall into local optimum, the Levy flight strategy is introduced to further improve the overall search ability of whale optimization algorithm, after the whale completes all the position update, a Levy flight is carried out to further update the position, the update method is as formula (29):

[0137] x(n+1)=x(n)+Levy(d)x(n) (29)

[0138] Wherein, n is the iteration number, d is the dimension of the position vector, wherein the expression of Levy(·) is:

[0139]

[0140] Wherein r1, r2 = rand(-1, 1), β = 1.5 is a constant, the expression of σ is as follows:

[0141]

[0142] Wherein, Gamma function.

[0143] The flow of L-SA-IWOA algorithm is shown in Figure 6 , and the specific steps are:

[0144] (1) initialize the value of each parameter, determine the objective function, generate random population of independent variable x;

[0145] (2) calculate the objective function value F(x), and record the optimal function value and the corresponding position X(n), update the value of ω, A, C, p;

[0146] (3) when p < 0.5 (i.e. the first set value) and |A| < 1 (i.e. the second set value), the surrounding prey update position according to formula (15);

[0147] (4) when p < 0.5 and |A| ≥ 1, update the position according to formula (24);

[0148] (5) when p ≥ 0.5 and |A| ≥ 1, update the position according to formula (20);

[0149] (6) generate new population x', calculate the function value F'(x) of the objective function, and record the corresponding position;

[0150] (7) if F'(x) is better than F best , the new population whale position replaces the original whale position;

[0151] (8) if F'(x) is worse than F best , calculate P according to formula (27)inferior Decide whether to accept the new population whale position;

[0152] (9) Thaw operation: t = 0.99 * t;

[0153] (10) Levy flight is carried out, and the position of itself is updated;

[0154] (11) Determine whether the maximum iteration number N is reached max If not, return to step (2) to continue iteration calculation, if the maximum iteration number is reached, proceed to the next step;

[0155] (12) Output the global optimal value and the corresponding x position.

[0156] According to the above analysis, the derivation and calculation of the line of the transformer area are carried out, and finally the impedance matrix of each line in the transformer area is obtained:

[0157]

[0158] Where Z nT represents the branch impedance, n represents the number of transformer area branches, and T represents the length of time series.

[0159] S4: Calculate the power loss of each line of the distribution network by using the root mean square current method

[0160] According to the data collected by the power grid, in the case of knowing the power consumption of the user end and the voltage of the transformer secondary side, the loss power of each branch is calculated by using the root mean square current method through power flow calculation.

[0161] This method equates the actual loss generated in the conductor in a period of time to the loss generated by the root mean square current. The actual loss of the line is as follows:

[0162]

[0163]

[0164] Where, ΔW is the power loss (KW·h), T is the power consumption time (h), R is the line resistance (Ω), i(t) is the actual current (A), and I a is the root mean square current (A).

[0165] If only the power loss of the line is calculated, it is as follows:

[0166]

[0167] In formula (35), I a The method is as follows:

[0168]

[0169] wherein, I i is the line current at time t.

[0170] If the data provided does not contain the current, I i can be calculated from the voltage, active power and reactive power:

[0171]

[0172] wherein, P i is the active power at time t of the line, Q i is the reactive power at time t of the line, U i is the voltage at time t of the line.

[0173] The rate loss can be expressed as:

[0174]

[0175] The impedance matrix (32) obtained in step (3) and the voltage, active power and reactive power of each branch of the line are combined, and the power loss matrix of each branch in the line can be obtained by calculation as follows:

[0176]

[0177] wherein, ΔP represents the loss power, n represents the number of branch circuits of the transformer area, and T represents the length of the time sequence.

[0178] S5: data analysis and fault diagnosis

[0179] After the above calculation steps, the power loss value of each branch of the transformer area in formula (39) is obtained, and the presence or absence of abnormalities is determined by analyzing the power loss of each branch at each time, so as to achieve the purpose of real-time monitoring and fault diagnosis.

[0180] Embodiment 2

[0181] The embodiment 2 of the present application provides a low-voltage power grid fault diagnosis system based on an improved whale optimization algorithm, comprising:

[0182] A data acquisition module configured to acquire low-voltage power grid transformer area data;

[0183] A topology generation module configured to obtain the topology structure of the low-voltage power grid according to the low-voltage power grid transformer area data;

[0184] An impedance calculation module configured to calculate the impedance of each branch of the low-voltage power grid based on the obtained topology structure by using the whale optimization algorithm improved based on Levy flight and simulated annealing;

[0185] A power loss calculation module configured to calculate the power loss of each branch by using the root mean square current method according to the impedance of each branch.

[0186] The fault diagnosis module is configured to obtain a low-voltage power grid fault diagnosis result according to the branch power loss.

[0187] The working method of the system is the same as the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm provided in Embodiment 1, and will not be described here.

[0188] Embodiment 3

[0189] Embodiment 3 of the present application provides a computer readable storage medium, which stores a program, and the program is executed by a processor to realize the steps in the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in Embodiment 1 of the present application.

[0190] Embodiment 4

[0191] Embodiment 4 of the present application provides an electronic device, which includes a memory, a processor, and a program stored in the memory and executable on the processor, and the processor executes the program to realize the steps in the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in Embodiment 1 of the present application.

[0192] The above only describes the preferred embodiments of the present application and is not used to limit the present application. For those skilled in the art, the present application can have various modifications and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A low-voltage power grid fault diagnosis method based on an improved whale optimization algorithm, characterized in that, The process includes the following: Acquire low-voltage power grid distribution area data; The topology of the low-voltage power grid is obtained based on the data from the low-voltage power grid distribution area. Based on the obtained topology, the impedance of each branch of the low-voltage power grid is calculated using the whale optimization algorithm based on Levy flight and simulated annealing. Based on the impedance of each branch, the power loss of each branch is calculated using the root mean square current method. Based on the power loss of each branch, the low-voltage power grid fault diagnosis results are obtained; The impedance of each branch of the low-voltage power grid was calculated using the whale optimization algorithm, which is based on Levy flight and simulated annealing. Based on the defined objective function, a random population of independent variables is generated; Calculate the objective function value, record the optimal function value and its corresponding position, and update the values ​​of the first matrix vector coefficients, the second matrix vector coefficients, the weighting factor, and the probability of the whale changing its behavior. When the probability of a whale changing its behavior is less than a first set value and the coefficient of the first matrix vector is less than a second set value, the whale will be surrounded and preyed upon to update its position. When the probability of a whale changing its behavior is less than a first set value and the coefficient of the first matrix vector is greater than or equal to a second set value, a random search is performed to update the position. When the probability of a whale changing its behavior is greater than or equal to a first preset value and the coefficient of the first matrix vector is greater than or equal to a second preset value, the position is updated in a spiral manner. A new population is generated, the updated objective function value is calculated, and the corresponding position is recorded; If the updated objective function value is better than the optimal value, then the new population whale position replaces the original whale position. If the updated objective function value is worse than the optimal value, calculate the probability of accepting the inferior solution to determine whether to accept the new population of whales' positions; Perform a cooling operation, initiate Levy flight, and update its own position; Determine if the maximum number of iterations has been reached. If not, return to continue iterative calculation. If the maximum number of iterations has been reached, proceed to the next step. Output the global optimal value and the corresponding position of the independent variable, and then obtain the matrix of impedance of each branch in the transformer area; The coefficients of the first matrix A are A = 2ar1 - a, and the coefficients of the second matrix C are C = 2r2, where a = 2 - 2n / Nmax, n is the number of iterations, r1 and r2 are random numbers between [0, 1], a is a coefficient that decreases linearly with the number of iterations, and Nmax represents the maximum number of iterations; Spiral Renewal Phase: After identifying prey, the population shrinks and the best individuals in the population spiral towards their optimal positions. The spiral update process is as follows: x ( n +1) = X ( n )+ω* D 2* ebl *cos(2π l ); ； in, D 2 represents the distance between the current position and the optimal position; b represents the spiral shape coefficient; l A random number in the range [-1, 1]. ω This represents the weighting factor, and k is the adjustment coefficient. x ( n+1 () represents the position of the whale in the (n+1)th iteration; X ( n ) represents the position of the optimal solution in the nth iteration.

2. The low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in claim 1, characterized in that, Low-voltage power grid distribution area data should include at least the voltage, current, active power, and reactive power of each branch in the distribution area.

3. The low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in claim 1, characterized in that, After acquiring the low-voltage power grid distribution area data, preprocessing of the low-voltage power grid distribution area data is performed. The preprocessing includes at least the following: using a Bayesian linear regression model to fill in the low-voltage power grid area data.

4. The low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in claim 1, characterized in that, Using the minimum mean standard deviation across all time dimensions as the objective function, the whale optimization algorithm, improved based on Levy flight and simulated annealing, is employed to calculate the impedance of each branch of the low-voltage power grid. The standard deviation is used to represent the difference between the calculated value and the actual value of the transformer secondary voltage derived from a certain user's data. Based on the standard deviation of all users at the same time, the mean function of the standard deviation under all time dimensions is obtained, and the minimum of this function is taken as the objective function.

5. The low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in claim 1, characterized in that, Based on the impedance of each branch, the power loss of each branch is calculated using the root mean square current method, including: Based on the impedance matrix of each branch in the transformer area, as well as the voltage, active power, and reactive power of each branch, the power loss matrix of each branch is obtained.

6. A low-voltage power grid fault diagnosis system based on an improved whale optimization algorithm, characterized in that, Implementing the steps in the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in any one of claims 1-5 includes: The data acquisition module is configured to acquire low-voltage power grid distribution area data. The topology generation module is configured to obtain the topology of the low-voltage power grid based on the low-voltage power grid distribution area data. The impedance calculation module is configured to calculate the impedance of each branch of the low-voltage power grid based on the obtained topology using the whale optimization algorithm improved by Levy flight and simulated annealing. The power loss calculation module is configured to calculate the power loss of each branch using the root mean square current method based on the impedance of each branch. The fault diagnosis module is configured to obtain low-voltage power grid fault diagnosis results based on the power loss of each branch.

7. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program implements the steps in the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in any one of claims 1-5.

8. An electronic device comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the low-voltage power grid fault diagnosis method based on the improved whale optimization algorithm as described in any one of claims 1-5.

Images