Novel power distribution network fault distance measurement method based on improved whale algorithm

Abstract

The invention relates to the technical field of power distribution network fault distance measurement, and discloses a novel power distribution network fault single-phase grounding distance measurement method based on an improved whale algorithm, and the method comprises the following steps: 1, constructing a distributed photovoltaic power distribution network model and a fitness function; step 2, unitizing a fault line by using a distributed parameter principle and a uniform transmission line principle, and calculating fault voltage and current before and after each point; and step 3, inputting the fault signal into a computer, and calculating a fitness function by using an improved whale algorithm so as to obtain a distance measurement result. The experimental conclusion proves the feasibility and accuracy of the method.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network fault location measurement, and in particular to a novel distribution network fault single-phase ground fault location measurement method based on an improved whale algorithm. Background Art

[0002] Electricity used by end users is transmitted from the transmission grid to the distribution network, an integral part of the power system. Accelerated urbanization and growing electricity demand are increasing the load on the distribution network, placing a direct impact on both residential life and industrial production. As distribution networks become increasingly overloaded due to rapid urbanization and rising electricity demand, environmental factors, aging equipment, and operational errors can easily lead to various faults, such as short circuits and ground faults. These faults disrupt power supply, impacting user access, potentially damaging equipment, posing safety risks, and ultimately resulting in economic losses. Statistics show that the probability of operational failure in the distribution system is approximately 80%. When a distribution system fault occurs, if the faulty line cannot be quickly and accurately disconnected, the stability of the distribution network will be compromised, and in severe cases, the system may even collapse. Therefore, finding a method for accurately and rapidly locating faults in distribution networks has become a crucial issue for ensuring the safe and stable operation of the power grid. Accurately locating the fault location lays the essential foundation for subsequent fault isolation, analysis, and power restoration, effectively reducing the economic losses caused by power outages. In this context, in-depth research into distribution network fault location is of great significance, offering both substantial economic and significant social benefits. Currently, power production and research institutions both domestically and internationally have been deeply attracted to this research field, and all parties involved are investing significant efforts in it. Summary of the Invention

[0003] (1) Technical problems solved

[0004] In view of the shortcomings of the existing technology, the present invention provides a new distribution network fault single-phase ground fault distance measurement method based on the improved whale algorithm, which solves the problems raised in the above background technology.

[0005] (2) Technical solution

[0006] To achieve the above object, the present invention provides the following technical solution: a new single-phase ground fault distance measurement method for distribution network based on an improved whale algorithm, comprising the following steps:

[0007] Step 1: Construct a distributed photovoltaic distribution network model and fitness function;

[0008] Step 2: Using the distributed parameter principle and the uniform transmission line principle, the fault line is unitized and the fault voltage and current before and after each point are calculated;

[0009] Step 3: Input the fault signal into the computer, and then use the improved whale algorithm to calculate the fitness function to obtain the ranging result.

[0010] Preferably, the detailed steps of step 3 are:

[0011] Step 31: Surround the prey. When hunting, the first thing a whale does is to determine the location of the prey, and then surround it. Once the prey's location is successfully determined, the other whales in the group will update their own positions based on the prey's current location. The expression is as follows:

[0012]

[0013] in, represents the position vector of a single whale, represents the best-fitting whale position vector now;

[0014] Step 32: Bubble predation uses two mechanisms to simulate bubble attack behavior: contraction update and spiral trajectory update. Through these two update methods, the position change of the bubble predation phase is completed. The expression is as follows:

[0015]

[0016] in, represents the distance between the humpback whale's location and the current optimal position; b represents the update constant when changing the spiral position;

[0017] Step 33: A threshold selection mechanism is introduced into the algorithm to achieve real-time synchronization of position changes. When the probability satisfies P < 0.5, the contraction process is skipped. If this condition is not met, a spiral update operation is performed. This is used to select the whale position update mechanism. The mathematical model for updating the position during the bubble attack phase is as follows:

[0018]

[0019] Where P is a random vector in [0,1];

[0020] Step 34: Searching for prey. In the whale algorithm's simulation environment, each whale's search location is unique and its search behavior is independent. While searching for prey, the whale's position changes in real time based on the coefficient A. However, once the absolute value of A exceeds 1, the other whales must move away from their current reference position. The mathematical model for the foraging phase is as follows:

[0021]

[0022] in, represents the position vector of a randomly selected whale;

[0023] Step 35: Improve the nonlinear convergence factor. Parameter A acts as a check and balance between local and global search. The value of A will change according to changes in relevant factors, and a1 decreases linearly from 2 to 0. This paper proposes a convergence factor that changes nonlinearly, and its formula is as follows:

[0024]

[0025] At the beginning, the slopes of a1 and a2 are relatively gentle, which helps to better avoid falling into the local optimum. In the middle and late stages of development, the slopes of a1 and a2 curves increase sharply, indicating that it has a stronger ability in local search;

[0026] Step 36: Introduce adaptive inertia weights. When other whales gradually approach the local optimal solution, the weight value will decrease accordingly. This helps improve the algorithm's ability to search for the optimal solution in the area near the optimal solution. The weight value and improved formula are as follows:

[0027]

[0028] Step 37: Binary processing of the whale algorithm. When using the binary whale optimization algorithm to solve the distribution network fault location problem, the continuous whale positions must be binary processed. The sigmoid function is used to compress the whale positions to the [0,1] interval, and then determine whether this position can become the new position. Let the conversion function be:

[0029]

[0030] Whale's location Update with 0 or 1:

[0031]

[0032] Among them, rand represents a random number uniformly distributed in [0,1]; Indicates the position of the whale after the binary update.

[0033] Preferably, the threshold selection mechanism is determined according to the situation when the algorithm is running normally.

[0034] Preferably, the fitness function value is determined according to the positive sequence voltage before and after the fault point.

[0035] Preferably, the novel single-phase ground fault ranging method for distribution network based on the improved whale algorithm is characterized in that the distribution network fault ranging parameters adopt distributed parameters and the uniform transmission lines are unitized to extract the voltage and current components before and after the fault point.

[0036] Preferably, the novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm is characterized in that the fault adopts a three-phase voltage module and the fault type is a single-phase ground fault.

[0037] Preferably, the novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm is characterized in that the fault line includes two types of lines: cable lines and overhead lines, and simulates a load-containing uniform transmission circuit and a distributed photovoltaic uniform transmission circuit.

[0038] Preferably, before performing the distance measurement, the method further includes filtering the collected voltage and current data to remove harmonic interference.

[0039] (3) Beneficial effects

[0040] Compared with the prior art, the present invention provides a new single-phase ground fault distance measurement method for distribution network based on an improved whale algorithm, which has the following beneficial effects:

[0041] This new single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm has a simple judgment structure and is easy to implement. The accuracy of the detection results of the present invention is not affected by fault factors such as fault distance, photovoltaic module temperature, fault closing angle, or grounding point transition impedance. It can quickly and accurately determine the fault point and fault distance for single-phase ground faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 Distributed parameter model diagram;

[0043] Figure 2 New distribution network fault equivalent topology diagram;

[0044] Figure 3 Whale algorithm flow chart;

[0045] Figure 4 Diagram of a 12-node simulation model with distributed photovoltaics;

[0046] Figure 5-Figure 8 Embodiment 1 of the present invention provides the distance measurement results of single-phase ground short circuit fault in a new distribution network based on the improved whale algorithm under different external conditions;

[0047] Figure 9 Embodiment 2 of the present invention provides ranging results of the improved whale algorithm and the original whale algorithm at different fault distances;

[0048] Figure 10 Embodiment 3 of the present invention provides the ranging results of the improved whale algorithm, gray wolf algorithm, and binary particle swarm algorithm at different fault distances; DETAILED DESCRIPTION

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0050] See also Figures 1-10 , a new distribution network fault location method based on an improved whale algorithm, the method comprising the following steps:

[0051] Step 1: Construct a distributed photovoltaic distribution network model and fitness function.

[0052] Step 2: Using the distributed parameter principle and uniform transmission line principle, the fault line is unitized and the fault voltage and current before and after each point are calculated.

[0053] Step 3: Input the fault signal into the computer, and then use the whale algorithm to calculate the fitness function to obtain the ranging result. This step includes the following specific judgment process:

[0054] Step 3.1: Surround the prey. When hunting, whales first need to determine the location of their prey before surrounding it. Once the prey's location is successfully determined, the other whales in the group will update their own positions based on the prey's current location. The expression is as follows:

[0055]

[0056] in, represents the position vector of a single whale, Represents the current best-fit whale position vector.

[0057] Step 3.2: Bubble predation uses two mechanisms to simulate bubble attack behavior: contraction update and spiral trajectory update. Through these two update methods, the position change of the bubble predation phase is completed. The expression is as follows:

[0058]

[0059] in, represents the distance between the humpback whale's location and the current optimal position; b represents the update constant when changing the spiral position.

[0060] Step 3.3: The algorithm introduces a threshold selection mechanism to achieve real-time synchronization of position changes. When the probability satisfies P < 0.5, the contraction process is skipped. If this condition is not met, a spiral update operation is performed. This is used to select the whale position update mechanism. The mathematical model for updating the position during the bubble attack phase is as follows:

[0061]

[0062] Where P is a random vector in [0,1].

[0063] Step 3.4: Searching for prey. In the whale algorithm's simulation environment, each whale's search location is unique and its search behavior is independent. While searching for prey, the whale's position changes in real time based on the coefficient A. However, once the absolute value of A exceeds 1, the other whales must move away from their current reference position. The mathematical model for exploring the foraging phase is as follows:

[0064]

[0065] in, represents the position vector of a randomly selected whale.

[0066] Step 3.5: Improve the nonlinear convergence factor. Parameter A acts as a check and balance between local and global search. The value of A will change based on changes in relevant factors, and a1 decreases linearly from 2 to 0. This paper proposes a convergence factor that changes nonlinearly, and its formula is as follows:

[0067]

[0068] At the beginning, the slopes of a1 and a2 are relatively gentle, which helps to better avoid falling into the local optimum. In the middle and late stages of development, the slopes of a1 and a2 curves increase sharply, indicating that it has a stronger ability in local search.

[0069] Step 3.6: Introduce adaptive inertia weights. When other whales gradually approach the local optimal solution, the weight value will decrease accordingly. This helps improve the algorithm's ability to search for the optimal solution in the area near the optimal solution. The weight value and improved formula are as follows:

[0070]

[0071] Step 4: Binary processing of the whale algorithm. When using the binary whale optimization algorithm to solve the distribution network fault location problem, the continuous whale positions must be binary processed. The sigmoid function is used to compress the whale positions to the [0,1] interval, and then determine whether this position can become the new position. Let the conversion function be:

[0072]

[0073] Whale's location Update with 0 or 1:

[0074]

[0075] Among them, rand represents a random number uniformly distributed in [0,1]; Indicates the position of the whale after the binary update.

[0076] Step 5: While controlling a single variable, change the fault grounding point transition impedance, fault closing angle, PV module temperature, and fault distance to simulate the impact of external environmental factors on the distribution network.

[0077] Through the above steps, the method of the present invention can accurately determine the fault distance under the overhead line and cable line when a single-phase ground short circuit fault occurs;

[0078] Invention implementation example 1:

[0079] Based on the Matlab / Simulink simulation platform, a 12-node simulation model containing distributed photovoltaics was analyzed, different fault parameter conditions were set, and the fault point was identified through the above steps. The specific results are as follows: Figure 5-8 As shown in the figure, the load module voltage is 380V, the active power is 700kW, and the induced reactive power is 320kVar. The fault module is a three-phase fault module, the fault type is a single-phase ground fault, the fault starts at 0.1s, and the fault line is a 15km overhead line or cable. This will not be further described.

[0080] Depend on Figure 5-Figure 8 As can be seen, single-phase ground faults have different impacts on distribution network fault location at different fault severity levels. However, despite the varying fault conditions, the distribution network fault location method proposed in this paper still demonstrates good accuracy under different fault severity levels, demonstrating that this method is unaffected by fault type or severity.

[0081] Invention implementation example 2:

[0082] Based on the Matlab / Simulink simulation platform, a 12-node simulation model with distributed photovoltaics was analyzed. By changing the distance at which the fault occurred, the ranging errors of the improved whale algorithm and the original whale algorithm were compared. The specific results are as follows: Figure 9 shown.

[0083] Depend on Figure 9 As can be seen, both IWOA and WOA perform well across a range of fault distances. However, the improved Whale algorithm achieves a lower fault error, with ranging errors within 2%. However, the original Whale algorithm suffers from poor stability, with errors potentially exceeding the 2% range. Therefore, the improved Whale algorithm is more suitable for the new distribution network model.

[0084] Invention implementation example 3:

[0085] Based on the Matlab / Simulink simulation platform, a 12-node simulation model with distributed photovoltaics was analyzed. By changing the distance at which the fault occurred, the ranging errors of the improved whale algorithm, the gray wolf algorithm, and the binary particle swarm algorithm were compared. The specific results are as follows: Figure 10 shown.

[0086] Depend on Figure 10 It can be seen that all three algorithms can meet the distance measurement requirements and can be used as new methods for single-phase grounding fault distance measurement in distribution networks. Relatively speaking, the relative error fluctuation of IWOA is smaller and its stability is better.

[0087] In summary, the detection method of the present invention can quickly and effectively measure the fault distance under different fault factors, has high accuracy, reliability and robustness, and has certain engineering application value.

[0088] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A new single-phase ground fault location method for distribution network based on improved whale algorithm, characterized by: The steps include: Step 1: Construct a distributed photovoltaic distribution network model and fitness function; Step 2: Using the distributed parameter principle and the uniform transmission line principle, the fault line is unitized and the fault voltage and current before and after each point are calculated; Step 3: Input the fault signal into the computer, and then use the improved whale algorithm to calculate the fitness function to obtain the ranging result.

2. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 1 is characterized by: The detailed steps of step 3 are: Step 31: Surround the prey. When hunting, the first thing a whale does is to determine the location of the prey, and then surround it. Once the prey's location is successfully determined, the other whales in the group will update their own positions based on the prey's current location. The expression is as follows: in, represents the position vector of a single whale, This represents the best-fitting whale position vector now: Step 32: Bubble predation uses two mechanisms to simulate bubble attack behavior: contraction update and spiral trajectory update. In each algorithm iteration, these two mechanisms are used to continuously update the local optimal solution. The expression is as follows: in, represents the distance between the humpback whale's location and the current optimal position; b represents the update constant when changing the spiral position; Step 33: A threshold selection mechanism is introduced into the algorithm to achieve real-time synchronization of position changes. When the probability satisfies P < 0.5, the contraction process is skipped. If this condition is not met, the spiral update operation is performed. The mathematical model for updating the position during the bubble attack phase is as follows: Where P is a random vector in [0,1]; Step 34: Searching for prey. In the whale algorithm's simulation environment, each whale's search location is unique and its search behavior is independent. While searching for prey, the whale's position changes in real time based on the coefficient A. However, once the absolute value of A exceeds 1, the other whales must move away from their current reference position. The mathematical model for the foraging phase is as follows: in, represents the position vector of a randomly selected whale; Step 35: Improve the nonlinear convergence factor. Parameter A acts as a check and balance between local and global search. The value of A will change according to changes in relevant factors, and a1 decreases linearly from 2 to 0. This paper proposes a convergence factor that changes nonlinearly, and its formula is as follows: At the beginning, the slopes of a1 and a2 are relatively gentle, which helps to better avoid falling into the local optimum. In the middle and late stages of development, the slopes of a1 and a2 curves increase sharply, indicating that it has a stronger ability in local search; Step 36: Introduce adaptive inertia weights. When other whales gradually approach the local optimal solution, the weight value will decrease accordingly. This helps improve the algorithm's ability to search for the optimal solution in the area near the optimal solution. The weight value and improved formula are as follows: Step 37: Binary processing of the whale algorithm. When using the binary whale optimization algorithm to solve the distribution network fault location problem, the continuous whale positions must be binary processed. The sigmoid function is used to compress the whale positions to the [0,1] interval, and then determine whether this position can become the new position. Let the conversion function be: Whale's location Update with 0 or 1: Among them, rand represents a random number uniformly distributed in [0,1]; Indicates the position of the whale after the binary update.

3. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 2 is characterized in that: The threshold selection mechanism is determined according to the situation when the algorithm is running normally.

4. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 3 is characterized by: The fitness function value is determined according to the positive sequence voltage before and after the fault point.

5. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 4 is characterized in that: The distribution network fault location parameters adopt distributed parameters and unitize the uniform transmission line to extract the voltage and current components before and after the fault point.

6. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 5 is characterized in that: The fault uses a three-phase voltage module and the fault type is a single-phase grounding fault.

7. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 6 is characterized in that: The fault lines include cable lines and overhead lines, and simulate the uniform transmission circuit containing loads and the uniform transmission circuit containing distributed photovoltaics.

8. The novel single-phase ground fault distance measurement method for distribution network based on the improved whale algorithm according to claim 7 is characterized in that: Before performing distance measurement, the method also includes a step of filtering the collected voltage and current data to remove harmonic interference.

Images