Photovoltaic power distribution network-containing single-phase earth fault distance measurement method based on cat swarm algorithm

Abstract

The invention relates to the technical field of power system fault detection, in particular to a photovoltaic power distribution network single-phase earth fault distance measurement method based on a cat swarm algorithm, and the method comprises the following steps: 1, building a distributed parameter model; step 2, constructing a fault model; step 3, setting a fitness function; according to the method, the cat swarm algorithm is combined with a distribution parameter distance measurement principle, and the accuracy and reliability of single-phase earth fault distance measurement of the photovoltaic power distribution network are improved by utilizing the advantages that the cat swarm algorithm is rapid in convergence and not prone to falling into a local optimal solution, so that the accuracy and reliability of fault distance measurement of the single-phase earth fault of the photovoltaic power distribution network are improved. Accurate results can be obtained in different fault conditions and environment changes, the method is not affected by distributed photovoltaic access, and the method has good universality.

Description

Technical Field

[0001] This invention relates to the field of power system fault detection technology, and in particular to a method for locating single-phase grounding faults in photovoltaic distribution networks based on the cat swarm algorithm. Background Technology

[0002] With the increasing scarcity of traditional energy sources, distribution networks containing distributed generation (such as photovoltaic power generation) are widely used. In such distribution networks, the integration of distributed generation makes the grid topology and current flow more complex. Traditional fault location methods are difficult to adapt to these changes and cannot meet the needs of complex grid structures and power flow distributions, leading to difficulties in fault location.

[0003] Currently, the main methods for fault location in distribution networks are impedance ranging and traveling wave ranging. Impedance ranging assumes a uniform impedance distribution in the power lines. It measures the three-phase current and voltage values ​​on both sides of the faulty distribution line and uses the distance between the fault location and the measured signal point, along with the proportional relationship between impedance and conductor length, to estimate the fault location. However, this method involves complex data processing and has limitations when considering distributed power source integration. Traveling wave ranging utilizes the time-series difference between the traveling wave dynamics generated at the fault location and the distribution line to determine the fault location. However, this method is highly sensitive to the phase angle at the moment of fault occurrence; at low phase angles, the traveling wave signal strength is weak, compromising the accuracy of the ranging results. Furthermore, the complex structure of distribution networks and the high cost of deploying high-rate sampling ranging equipment limit its widespread application. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] To address the shortcomings of existing technologies, this invention provides a single-phase grounding fault location method for photovoltaic distribution networks based on the cat swarm algorithm, thereby improving the accuracy and reliability of fault location and meeting the actual needs of photovoltaic distribution networks.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution: a method for single-phase ground fault location in photovoltaic distribution networks based on the cat swarm algorithm, comprising the following steps:

[0008] Step 1: Establish a distributed parameter model: Taking into account the distribution characteristics of line resistance, inductance, and capacitance in photovoltaic distribution networks and the location of photovoltaic access, the uniform transmission line is divided into small units using the segmented micro-element method. Based on Kirchhoff's laws, a distributed parameter model that can accurately describe the distribution of line voltage and current is established.

[0009] Step 2: Constructing a fault model: Taking a uniform distribution line with a photovoltaic distribution network as the object, and considering the single-phase grounding fault type, a mathematical relationship model between the voltage / current at the fault point and the electrical quantities at both ends of the line is established, taking into account factors such as the grounding resistance at the fault point and the time of occurrence.

[0010] Step 3: Set the fitness function: Based on the actual needs such as the distribution network topology and fault types, design a fitness function with the voltage across the fault location as the core parameter to measure the accuracy of the candidate fault location;

[0011] Step 4: Use the cat swarm algorithm for fault location: By adjusting the algorithm configuration items (such as memory pool, dimensional change domain, etc.), clone and randomly perturb the cat swarm position in search mode, and combine it with the position update based on the global optimal solution in tracking mode to iteratively optimize the fitness function and locate the optimal fault location.

[0012] Preferably, the detailed steps of step 4 are as follows:

[0013] Step 41: Adjust algorithm configuration items: Search memory pool (SMP), Self-positioning judgment (SPC), Dimension change range (SRD), Dimension change number (CDC);

[0014] Step 42: Copy the cat's location: Copy the cat's location j times, perform the SPC process. If SPC is true, j = SMP - 1; otherwise, j = SMP, and store the copy in SMP.

[0015] Step 43: Implement random perturbation: Apply a random perturbation to each clone based on the CDC and SDR values ​​to obtain a new location;

[0016] Step 44: Evaluation and Location Update: Based on the functional performance of each clone within the SMP, evaluate the suitability of the new location point within the SMP, trace and locate the clone corresponding to the best functional value, and use it to update the cat's current location in the search mode.

[0017] Preferably, the tracking mode in the cat swarm algorithm is a motion mode that simulates a cat hunting, similar to the particle swarm algorithm. It uses the global optimal position to update the cat's current speed, and then updates the cat's current position by iterating the new speed in turn, so that the cat's movement continuously tends to the optimal solution.

[0018] Preferably, before using the cat swarm algorithm for fault location, a preprocessing step is included for the acquired electrical signals. The preprocessing includes filtering to remove noise interference and improve the quality of the electrical signals.

[0019] Preferably, the method further includes a step of verifying the fault location result after obtaining the fault location result. Specifically, this involves performing simulation verification by changing external environmental factors such as the grounding resistance of the fault point, ambient temperature, and fault distance to determine the accuracy and reliability of the location result.

[0020] Preferably, the form of the fitness function can be adjusted and optimized according to the characteristics and needs of the actual distribution network to better adapt to different fault location scenarios.

[0021] Preferably, the parameters of the cat swarm algorithm can be dynamically adjusted according to the scale and complexity of the power distribution network and the requirements for calculation speed and accuracy, so as to improve the performance of the algorithm.

[0022] (III) Beneficial Effects

[0023] Compared with existing technologies, this invention provides a method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm, which has the following advantages:

[0024] This invention combines the cat swarm algorithm with the principle of distributed parameter ranging. It utilizes the advantages of the cat swarm algorithm, such as fast convergence and low susceptibility to local optima, to improve the accuracy and reliability of single-phase ground fault ranging in photovoltaic distribution networks. It can obtain accurate results under different fault conditions and environmental changes, is unaffected by distributed photovoltaic access, and has good universality.

[0025] Compared to traditional optimization algorithms, the cat swarm optimization algorithm possesses both superior global search capabilities and convergence speed, effectively improving computational speed. Compared to the particle swarm optimization algorithm, it requires fewer iterations, has a faster computation speed, and provides more reliable and ideal results.

[0026] This invention uses a fitness function to measure the feasibility and accuracy of candidate fault locations, enabling more precise fault location determination. Simultaneously, the search and tracking modes of the cat swarm algorithm are designed in detail to better adapt to the needs of fault location in distribution networks.

[0027] The acquired electrical signals are preprocessed to remove noise interference, improving signal quality and thus enhancing fault location accuracy. Verification of the ranging results ensures the reliability of this method in practical applications. Attached Figure Description

[0028] Figure 1 This is a diagram of the photovoltaic power generation grid connection model in this invention;

[0029] Figure 2 This is the distributed parameter model in this invention;

[0030] Figure 3This invention relates to the equivalent topology of a distributed photovoltaic power distribution network with faults.

[0031] Figure 4 This is a flowchart of the cat group search algorithm in this invention;

[0032] Figure 5 This invention includes a fault simulation circuit for distributed photovoltaic power distribution networks;

[0033] Figure 6 This invention compares the convergence changes of the cat swarm algorithm and the particle swarm algorithm during fault location. Detailed Implementation

[0034] 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.

[0035] Please see Figures 1-6 The flowchart of the single-phase grounding fault location method for photovoltaic distribution networks based on the cat swarm algorithm of the present invention is shown in the figure. The method includes the following steps:

[0036] Step 1: Establish a distributed parameter model, from Figure 2 It can be seen that, considering the distribution characteristics of parameters such as resistance, inductance, and capacitance of lines in photovoltaic distribution networks, as well as the access location and output characteristics of distributed photovoltaic systems, the uniform transmission line is subdivided into multiple tiny units using the concept of segmented infinitesimal elements. Based on the fundamental laws of circuits, Kirchhoff's Current Law (KCL) and Kirchhoff's Voltage Law (KVL), the formulas are as follows:

[0037]

[0038]

[0039] in, These represent the initial voltage and current of the circuit unit, respectively. dx represents the back-end voltage and current of the circuit unit, respectively; dx represents the length of the circuit unit; L0dx, C0dx, G0dx, and R0dx are the electrical parameters of the circuit unit.

[0040] The impedance and admittance of the overhead line unit are Z1 and Y1, respectively, and those of the cable unit are Z2 and Y2. The angular frequency is ω, α is the attenuation constant, and β is the phase constant. Then, the expressions for the propagation coefficients of the overhead line and the cable are:

[0041]

[0042] The expression for characteristic impedance is:

[0043]

[0044] Calculate the voltage and current equations for any distance x along a uniform transmission line:

[0045]

[0046] Establish a distributed parameter model that accurately reflects the electrical characteristics of the distribution network. This model can precisely describe the distribution of voltage and current on the lines, providing a reliable basis for subsequent fault analysis.

[0047] Step 2: Construct a fault model, such as Figure 5 This study focuses on uniform distribution lines with distributed photovoltaic (PV) distribution networks, and constructs a corresponding fault model for single-phase grounding faults, a common fault type. During model construction, the influence of factors such as the grounding resistance at the fault point and the time of fault occurrence on the fault current and voltage is fully considered. Through analysis of the line distributed parameter model and combined with fault conditions, mathematical relationships between the voltage and current at the fault point and the voltage and current at both ends of the line are established, providing a theoretical basis for fault location.

[0048] Step 3: Set the fitness function. Based on the fault location objective and the actual conditions of the distribution network, design a reasonable fitness function. The fitness function is used to measure the quality of each candidate solution (i.e., possible fault locations) in the cat swarm algorithm. Considering the complexity and diversity of actual distribution networks, the form of the fitness function can be adjusted and optimized according to factors such as the distribution network topology, load characteristics, and fault types. For example, parameters related to fault current and voltage can be introduced into the fitness function, and the fitness of candidate solutions can be evaluated by calculating the error between the measured values ​​and the model calculated values. Furthermore, the weights of each parameter in the fitness function can be adjusted according to different fault types and distribution network operating states to better adapt to different fault location scenarios and improve the accuracy of fault location. The fitness function used in this paper is shown in the formula:

[0049]

[0050] In the formula, f0 represents the expected optimal fitness. and The voltage across the fault location is given in the formula.

[0051] Step 4: Use the cat swarm algorithm for fault location. The cat swarm search process is as follows: Figure 4 .

[0052] Step 4.1: Adjust the algorithm configuration items: Search Memory Pool (SMP), Self-Positioning Judgment (SPC), Dimension Change Range (SRD), Dimension Change Count (CDC).

[0053] Step 4.2: Copy the cat's location: Copy the cat's location j times and perform the SPC process. If SPC is true, j = SMP - 1; otherwise, j = SMP and store the copy in SMP.

[0054] Step 4.3: Implement random perturbation: Apply a random perturbation to each clone based on the CDC and SDR values ​​to obtain a new position.

[0055] Step 4.4: Evaluation and Location Update: Based on the functional performance of each clone within the SMP, evaluate the suitability of the new location within the SMP, trace and locate the clone corresponding to the best functional value, and use it to update the cat's current location in the search mode.

[0056] Step 5: Electrical Signal Preprocessing. Before using the cat swarm algorithm for fault location, the acquired electrical signals are preprocessed, including filtering. Appropriate filtering algorithms, such as Kalman filtering or wavelet filtering, are used to remove noise interference and improve the quality of the electrical signals. High-quality electrical signals provide more accurate data for fault location, thereby improving the accuracy of fault location.

[0057] Step 6: Verification of Ranging Results. After obtaining the fault ranging results, the results are verified. Simulation verification is performed by changing external environmental factors such as the grounding resistance of the fault point, ambient temperature, and fault distance. Under different simulation conditions, the ranging results are compared with the actual fault location to determine the accuracy and reliability of the ranging results. For example, multiple simulations are conducted at different fault distances to observe the changes in ranging error; the grounding point transition impedance is changed to test the ranging performance of the algorithm under different fault resistance conditions; different ambient temperatures are simulated to evaluate the impact of temperature on distributed photovoltaic power output and fault ranging results. Figure 6 Furthermore, by comparing the method with the particle swarm optimization algorithm for fault location, and through comprehensive simulation verification, the reliability of this method in practical applications is ensured.

[0058] Step 7: Dynamically Adjust Algorithm Parameters. The parameters of the cat swarm algorithm can be dynamically adjusted according to the scale and complexity of the distribution network, as well as the requirements for computational speed and accuracy. For large-scale and complex distribution networks, appropriately increasing the population size and maximum number of iterations can improve the algorithm's search capability and ensure that it can find the global optimum. For scenarios with high computational speed requirements, parameters can be adjusted to accelerate the convergence speed while ensuring a certain level of accuracy. For example, in urban distribution networks, due to the complexity of lines and the large number of nodes, the population size can be increased to enable the algorithm to search for fault locations more comprehensively. In some industrial distribution networks with high real-time requirements, the maximum number of iterations can be appropriately reduced to improve computational speed.

[0059] Through the above steps, the method of the present invention can accurately measure the distance under a single-phase ground fault in a photovoltaic distribution network, and is not affected by distributed photovoltaic access, thus achieving a rapid response.

[0060] Invention Implementation Example 1:

[0061] Using existing simulation tools MATLAB / SIMULINK, a single-phase ground fault simulation model of a distribution network containing distributed generation (DG) was constructed. The system consists of a three-phase power supply, an overhead line uniform transmission circuit, a cable uniform distribution circuit, a single-phase ground fault, and photovoltaic power generation modules. A 400kW distribution network system simulation model was established, with the circuit consisting of four parallel uniform transmission lines. The model simulates uniform distribution circuits with loads, uniform distribution circuits with branch power sources, and uniform distribution circuits with distributed photovoltaics, respectively, to verify the versatility of the ranging method. All simulation models use distributed parameter models to make the simulation model more consistent with real-world conditions.

[0062] The load module voltage is set to 380V, the rated operating frequency is 50Hz, the active power is 700kW, and the induced reactive power is 320kVar. A three-phase fault module is used, with a single-phase ground fault type and a fault start time of 0.1s. The circuit parameters per unit length of the distribution line are shown in Table 1.

[0063] Invention Implementation Example 2:

[0064] The experimental environment was as follows: Intel(R) Core(TM) i5-9300HF CPU @ 2.40GHz, RAM 8.00GB, Windows 11 64-bit operating system, and MATLAB 2022a as the software simulation environment. The initial settings for the cat swarm algorithm parameters are shown in Table 2.

[0065] Invention Implementation Example 3:

[0066] Different simulation parameters were set in MATLAB, and the voltage and current values ​​at the beginning and end of the single-phase ground fault distribution line were measured under each simulation condition. The ranging results were obtained through the ranging model.

[0067] Simulation under different fault distances: The system experiences the same single-phase ground fault (both A-level ground faults), with a fault occurrence time of 0.1s, a fault closing angle of 0°, and a grounding point transition impedance of 100Ω. On a 20km length of cable or overhead line containing only load, the single-phase ground fault is connected to different fault distances for simulation ranging. The results show that the single-phase ground fault ranging method based on the CSO algorithm has very high accuracy in both cable and overhead line fault ranging. Under different fault distances, the error of the ranging method is basically no more than 1%, as shown in Table 3.

[0068] Simulation under different grounding point transition impedances: The system experiences the same single-phase ground fault (both are A-ground faults), with a fault occurrence time of 0.1s and a fault closing angle of 0. The fault grounding occurs on a 20km long cable or overhead line containing only load. The single-phase ground fault is connected to the fault at a distance of 10km. Different transition impedances are set at the grounding points, and simulated fault distance measurement is performed. The results show that this fault distance measurement method can still obtain qualified and correct distance measurement results under different fault impedances. Moreover, the distance measurement effect is even better when the fault resistance is large, and the fault error can be controlled within a very small range, as shown in Table 4.

[0069] Simulation under different fault closing angles: The system experiences the same single-phase ground fault (both are A-level ground faults). The fault grounding occurs on a 20km long cable line containing only the load. The single-phase ground fault connection distance is 10km. Different fault times are set at the grounding point to change the fault closing angle, and simulated fault distance measurement is performed. The results show that this fault distance measurement method can still obtain accurate distance measurement results under different fault closing angles, with an error much less than 10%, as shown in Table 5.

[0070] Simulation under different temperatures of distributed photovoltaic modules: A simulation model was constructed to simulate the daily operation of distributed photovoltaic modules. The system experienced the same single-phase ground fault (both A-level ground faults), with a fault occurrence time of 0.1s and a fault closing angle of 0. The fault grounding occurred on a 20km long cable containing only the load, and the single-phase ground fault connection distance was 10km. The transition impedance of the grounding point was set to 100Ω. The photovoltaic module temperature was varied from 10℃ to 50℃, with temperature settings every 5℃, and simulated distance measurement was performed. The results show that this fault distance measurement method can still obtain accurate distance measurement results under different photovoltaic module temperature environments, with an error of approximately 1%.

[0071] Simulation comparison with particle swarm optimization (PSO): Simulations were performed using both the CSO algorithm and the binary particle swarm optimization (BPSO) algorithm for fault location. The system experienced the same single-phase ground fault (both A-level ground faults), with a fault occurrence time of 0.1 s and a grounding point transition impedance of 100 Ω. The single-phase fault was connected to a 10 km section of a 20 km long cable or overhead line containing only load. Results show that the proposed algorithm and the BPSO algorithm have comparable location accuracy. However, the proposed algorithm is faster in fault location and less prone to getting trapped in local optima, achieving more accurate results with fewer iterations. Figure 6 As shown in Table 6.

[0072]

[0073]

[0074] Table 1

[0075]

[0076] Table 2

[0077]

[0078] Table 3

[0079]

[0080] Table 4

[0081]

[0082] Table 5

[0083]

[0084] Table 6

[0085] In summary, the above embodiments have verified the accuracy and effectiveness of the single-phase grounding fault location method for photovoltaic distribution networks based on the cat swarm algorithm of this invention. In practical applications, the algorithm parameters and fitness function can be further optimized according to the specific distribution network conditions to better meet engineering requirements.

[0086] In the description herein, it should be noted that relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0087] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm, characterized in that, Includes the following steps: Step 1: Establish a distributed parameter model: Taking into account the distribution characteristics of line resistance, inductance, and capacitance in photovoltaic distribution networks and the location of photovoltaic access, the uniform transmission line is divided into small units using the segmented micro-element method. Based on Kirchhoff's laws, a distributed parameter model that can accurately describe the distribution of line voltage and current is established. Step 2: Constructing a fault model: Taking a uniform distribution line with a photovoltaic distribution network as the object, and considering the single-phase grounding fault type, a mathematical relationship model between the voltage / current at the fault point and the electrical quantities at both ends of the line is established, taking into account factors such as the grounding resistance at the fault point and the time of occurrence. Step 3: Set the fitness function: Based on the actual needs such as the distribution network topology and fault types, design a fitness function with the voltage across the fault location as the core parameter to measure the accuracy of the candidate fault location; Step 4: Use the cat swarm algorithm for fault location: By adjusting the algorithm configuration items (such as memory pool, dimensional change domain, etc.), clone and randomly perturb the cat swarm position in search mode, and combine it with the position update based on the global optimal solution in tracking mode to iteratively optimize the fitness function and locate the optimal fault location.

2. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 1, characterized in that, The specific steps of the search pattern in the cat swarm algorithm are as follows: Step 41: Adjust algorithm configuration items: Search memory pool (SMP), Self-positioning judgment (SPC), Dimension change range (SRD), Dimension change number (CDC); Step 42: Copy the cat's location: Copy the cat's location j times and perform the SPC process. When SPC is true, j = SMP-1. Otherwise, if j = SMP, store the copy in SMP; Step 43: Implement random perturbation: Apply a random perturbation to each clone based on the CDC and SDR values ​​to obtain a new location; Step 44: Evaluation and Location Update: Based on the functional performance of each clone within the SMP, evaluate the suitability of the new location point within the SMP, trace and locate the clone corresponding to the best functional value, and use it to update the cat's current location in the search mode.

3. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 2, characterized in that: The tracking mode in the cat swarm algorithm simulates the movement pattern of a cat hunting, similar to the particle swarm algorithm. It uses the global optimal position to update the cat's current speed, and then updates the cat's current position by iterating the new speed in turn, so that the cat's movement continuously tends to the optimal solution.

4. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 3, characterized in that: Before using the cat swarm algorithm for fault location, a preprocessing step is included in the acquired electrical signal. The preprocessing includes filtering to remove noise interference and improve the quality of the electrical signal.

5. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 4, characterized in that: The method also includes a step of verifying the fault location results after obtaining them. Specifically, it involves performing simulation verification by changing external environmental factors such as the grounding resistance of the fault point, ambient temperature, and fault distance to determine the accuracy and reliability of the location results.

6. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 5, characterized in that: The form of the fitness function can be adjusted and optimized according to the characteristics and needs of the actual distribution network to better adapt to different fault location scenarios.

7. The method for single-phase grounding fault location in photovoltaic distribution networks based on the cat swarm algorithm according to claim 6, characterized in that: The parameters of the cat swarm algorithm can be dynamically adjusted according to the scale and complexity of the power distribution network, as well as the requirements for calculation speed and accuracy, in order to improve the performance of the algorithm.

Images