Power distribution network pilot protection method, device and system, and storage medium
By constructing equivalent circuit diagrams and PSCAD simulation models, and combining abnormal data repair and Hellinger distance discrimination methods, the accuracy and reliability issues of longitudinal protection in distribution networks under complex conditions were solved, achieving efficient fault diagnosis and protection.
Patent Information
- Application Number
- CN202610487258.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-14
- Publication Date
- 2026-07-03
AI Technical Summary
Traditional power distribution network protection methods struggle to accurately and reliably perform fault diagnosis and longitudinal protection under conditions of high noise, data anomalies, and communication delays, especially given the increasing penetration of distributed power sources and the integration of power electronic equipment. Furthermore, they are severely affected by electromagnetic interference and data loss.
Equivalent circuit diagrams and PSCAD simulation models were constructed to obtain fault response data. The forward difference method was used to identify abnormal current values. The abnormal data were repaired by cubic trigonometric Hermite interpolation with shape parameters. An improved voltage gradient algorithm was used to construct a start-up criterion. The Hellinger distance was calculated by combining kernel density estimation to realize the probability distribution of current at both ends of the faulty line. Finally, the fault within the faulty area was identified by the Hellinger distance.
It improves the data accuracy and reliability of longitudinal protection in distribution networks, enhances the sensitivity and speed of fault diagnosis, and has good prospects for engineering applications.
Smart Images
Figure CN122338684A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power distribution network relay protection technology, specifically relating to a power distribution network longitudinal protection method, device, system, and storage medium. Background Technology
[0002] With the continuous increase in the penetration rate of distributed power sources and the large-scale integration of power electronic equipment in distribution networks, the fault characteristics of distribution networks have undergone profound changes. Traditional overcurrent protection and single-ended quantity protection face severe challenges in adaptability, selectivity, and speed when dealing with bidirectional power flow, weak feed current, and complex fault characteristics. How to accurately and reliably achieve fault diagnosis and longitudinal protection of distribution network lines under adverse conditions such as high noise, data anomalies, and communication delays has become a key scientific problem in the field of distribution network protection. Meanwhile, in practical engineering applications, due to factors such as electromagnetic interference, transformer saturation, or synchronous measurement errors, the current data collected by protection devices inevitably contains abnormal values or missing data. If these are not identified and repaired, they will directly lead to misjudgment or failure to operate by the protection algorithm. Therefore, constructing a distribution network longitudinal protection method that combines abnormal data processing capabilities with high-sensitivity fault discrimination performance has significant theoretical and practical value for improving the reliability of distribution network power supply and ensuring the safe and stable operation of the power grid. Summary of the Invention
[0003] To address the problems existing in the prior art, this invention provides a method, device, system, and storage medium for longitudinal protection of distribution networks. In the abnormal data extraction and repair section, fault response data is obtained by constructing equivalent circuit diagrams and PSCAD simulation models. Current anomalies are identified using the forward and backward difference method, and the abnormal data is accurately repaired using cubic trigonometric Hermite interpolation with shape parameters. In the longitudinal protection method based on Hellinger distance, an improved voltage gradient algorithm is used to construct a start-up criterion, and kernel density estimation is used to calculate the probability distribution of currents at both ends of the faulty line, thereby solving for the Hellinger distance to construct the protection criterion. In the section on constructing a complete longitudinal protection method for distribution networks, the above data repair is combined with the Hellinger distance discrimination method. Through abnormal data detection and repair, Hellinger distance calculation, and threshold value comparison, fault discrimination and fault phase diagnosis within the area are achieved, ultimately completing the longitudinal protection of the distribution network.
[0004] To achieve the above objectives, the present invention provides the following solution: A method for longitudinal protection of a distribution network, comprising: Step 1: Extract and repair key response data that are strongly related to line protection; Step 2: Construct a longitudinal protection criterion based on Hellinger distance; Step 3: Implement longitudinal protection of the distribution network based on the key response data after abnormal data repair and the longitudinal protection criteria.
[0005] Preferably, in step S1, the abnormal current data is accurately repaired using cubic trigonometric Hermite interpolation with shape parameters.
[0006] As a preferred option, in step 2, an activation criterion is constructed using an improved voltage gradient algorithm, and the probability distribution of the current at both ends of the faulty line is calculated using kernel density estimation, thereby solving for the Hellinger distance to construct a protection criterion; in step 3, data repair is combined with Hellinger distance discrimination, and fault discrimination and fault phase diagnosis are achieved within the area through abnormal data detection and repair, Hellinger distance calculation and threshold value comparison, ultimately completing the longitudinal protection of the distribution network.
[0007] The present invention also provides a longitudinal protection device for a power distribution network, comprising: The first processing module is used to extract and repair key response data that is strongly related to line protection; The second processing module is used to construct longitudinal protection criteria based on Hellinger distance; The third processing module is used to implement longitudinal protection of the distribution network based on the key response data after abnormal data repair and the longitudinal protection criteria.
[0008] As a preferred option, the first processing module uses cubic trigonometric Hermite interpolation with shape parameters to accurately repair abnormal current data.
[0009] As a preferred embodiment, the second processing module constructs a start-up criterion through an improved voltage gradient algorithm and uses kernel density estimation to calculate the probability distribution of the current at both ends of the faulty line, thereby solving for the Hellinger distance to construct a protection criterion; the third processing module combines data repair with Hellinger distance discrimination, and realizes fault discrimination and fault phase diagnosis within the area through abnormal data detection and repair, Hellinger distance calculation and threshold value comparison, and finally completes the longitudinal protection of the distribution network.
[0010] The present invention also provides a distribution network longitudinal protection system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a distribution network longitudinal protection method when executed by the processor.
[0011] The present invention also provides a storage medium storing a computer program, which executes a power distribution network longitudinal protection method when running.
[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention addresses the challenges of ensuring data quality in distribution networks under complex operating conditions and the insufficient adaptability of traditional protection methods. It proposes a longitudinal protection method for distribution networks based on anomaly data repair and Heringer distance. Unlike traditional protection principles that rely solely on electrical quantity amplitude comparison or phase differential, this method first constructs an equivalent circuit model of a typical distribution network using a PSCAD simulation platform. This model simulates ground faults occurring at different locations and extracts current response data from both ends of the faulty line pole, constructing a key response feature dataset. Secondly, to identify potential outliers in the fault data, an anomaly data identification algorithm based on forward and backward differential is used to rapidly detect the current data at both ends of the faulty line. Furthermore, a cubic trigonometric Hermite interpolation method with shape parameters is introduced. By constructing interpolation spline curves that satisfy the second-order continuity condition, the identified anomaly data is accurately repaired, effectively eliminating the interference of data noise on subsequent fault identification. Then, based on data preprocessing, an improved voltage gradient algorithm is used to construct a protection initiation criterion and monitor line voltage changes in real time. After protection initiation, the probability distribution of the currents at both ends of the faulty line is fitted using a kernel density estimation method, and the Helling distance between them is calculated to quantify the similarity of the current waveforms at both ends. Finally, by setting a threshold value for the Helling distance, reliable discrimination between faults within and outside the protection zone is achieved, and the selection of the faulty phase and the identification of the fault type are further completed. Simulation results show that the proposed method can maintain high accuracy and reliability of protection actions even under abnormal data interference, and has good prospects for engineering applications. Attached Figure Description
[0013] To more clearly illustrate the technical solution of the present invention, the drawings used in the embodiments are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0014] Figure 1 This is a flowchart of the power distribution network longitudinal protection method according to an embodiment of the present invention; Figure 2 This is a topology diagram of a photovoltaic-connected distribution network based on abnormal data repair and Hellinger distance for longitudinal protection of the distribution network according to the present invention. Detailed Implementation
[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0016] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0017] Example 1 like Figure 1 , 2 As shown, the present invention provides a longitudinal protection method for a distribution network, comprising: 1) Extraction of key response data and repair of abnormal data that are strongly correlated with line protection; Step 1: Taking a typical power system as an example, construct an equivalent circuit diagram to simulate short-circuit faults in the actual system. Build a simulation model using the PSCAD platform, set a ground fault, and obtain response data for various branches at different fault locations. Step 2: Identification of abnormal current values at both ends of the faulty line pole, using the following formula: In the formula: and In order to be in Forward and backward differences at the point; This represents the error threshold for forward and backward difference.
[0018] Step 3: Use cubic triangular Hermite interpolation with shape parameters to identify and repair abnormal data in the longitudinal protection fault data. In the cubic triangular Hermite interpolation with shape parameters, within the interval [...] i b , i b+1 [Middle Note] h = ( ) / di Then the interval [ i b , i b+1 The basis functions on ] are about h (0≤ h For functions with ≤1), the interpolation formula is as follows: In the formula: To adjust the parameters.
[0019] Clearly, the above equation satisfies the following conditions: In the formula: when hour, ;when hour, , .
[0020] For the current data at both ends of the faulty line pole { i 1, i 2, ... i a}remember y b and z b They are the dividing points. i b Given the function value and first derivative at a given point, we have a cubic trigonometric Hermite interpolation spline curve: This interpolated spline curve can satisfy up to second-order continuity, and can be modified by changing... The value is used to change the shape of the interpolation curve, which improves the accuracy and universality of interpolation and can be used to repair the current data at both ends of the faulty line pole.
[0021] 2) A longitudinal protection method based on Hellinger distance; Step 1: The longitudinal protection method based on Hellinger distance is activated by detecting voltage changes on the faulty line. When a fault occurs on a distribution network line, the line voltage changes from an approximately constant state to a voltage surge state. Therefore, the occurrence of a fault can be determined by detecting voltage changes. An improved voltage gradient algorithm is used to construct the activation criterion: In the formula: This is the voltage gradient value, which is the difference between the sums of the three values before and after the sampling point; This is the start-up threshold value. The protection start-up threshold value must simultaneously meet two conditions: its value is less than the voltage gradient value during a fault and greater than the voltage gradient value during normal operation.
[0022] Step 2: After the protection system is activated, process the faulty line current data. The probability distribution of the current at both ends of the faulty line is as follows: p (i) and q (i) ,but: In the formula: the current at both ends of the faulty line is and ; h Set the window width to 1; K [□] For kernel functions, the Gaussian kernel function is generally used.
[0023] Step 3: Calculate the Hellinger distance between the currents at both ends of the faulty line. The calculation formula is as follows: In the formula: The Hellinger distance is the distance between the currents at both ends of the faulty line.
[0024] Step 4: The fault diagnosis is completed by using the Hellinger distance between the currents at both ends of the faulty line calculated above, which causes the circuit breaker to operate.
[0025] 3) Distribution network longitudinal protection method based on abnormal data repair and Hellinger distance.
[0026] Step 1: Record the voltage and current data at the protection measurement point.
[0027] Step 2: Determine whether a fault has occurred using a starting criterion based on an improved voltage gradient algorithm. If the voltage does not reach the threshold value, return to Step 1. Otherwise, the protection is activated, the activation time is recorded, and this recorded time is set as the start time. The fault line current is also recorded 5ms after the fault occurs, and this information is transmitted to the protection on the opposite side.
[0028] Step 3: Protection Scheme. The system detects abnormal data by comparing the data before and after the data. If abnormal data is detected, it is deleted and the abnormal data is repaired using cubic triangular Hermite interpolation with shape parameters.
[0029] Step 4: Calculate the Hellinger distance between the phase currents on both sides of the same phase of the faulty line based on the above data. If the Hellinger distances between phases a, b, and c exist... If the fault is within the zone, then it is an in-zone fault; otherwise, it is an out-of-zone fault.
[0030] Step 5: Determine the faulty phase based on the calculated Hellinger distance. If a phase meets the following conditions... If the fault occurs in that phase, the fault type can be diagnosed using the methods described above.
[0031] This invention discloses a longitudinal protection method for distribution networks, comprising: extraction of key response data strongly correlated with line protection and repair of abnormal data; a longitudinal protection method based on Hellinger distance; and a longitudinal protection method for distribution networks based on abnormal data repair and Hellinger distance. This method extracts key response data strongly correlated with line protection through abnormal data identification and repair, effectively solving the problem of missing or distorted fault data and improving data accuracy and reliability. This application utilizes cubic trigonometric Hermite interpolation with shape parameters to accurately repair abnormal current data, overcoming the limitations of poor adaptability and low accuracy of traditional data repair methods, and ensuring the effective preservation of fault characteristics. Simultaneously, this invention constructs a longitudinal protection criterion based on Hellinger distance, achieving accurate identification of faults within and outside the fault zone through in-depth mining of the current probability distribution at both ends of the faulty line. Compared with traditional longitudinal protection methods, it has significant advantages in fault diagnosis sensitivity, speed, and resistance to abnormal data interference, and is scientifically sound and highly applicable in engineering.
[0032] Example: The specific process of the distribution network longitudinal protection method based on abnormal data repair and Hellinger distance proposed in this invention is as follows: Figure 1 As shown in the figure. The effectiveness of the proposed method is verified in the distribution network system after photovoltaic (PV) integration. The distribution network system diagram after PV integration is shown in the figure. Figure 2 As shown in the figure, two-phase short-circuit ground faults (ab, bc, ac, and abc) were set at 50% of the total length of the distribution network between busbar A and busbar B (hereinafter referred to as line A-B), respectively, within the zone. Simulation verification was performed by setting a three-phase short-circuit ground fault (abc) outside the zone. The protection action results for faults inside and outside the zone are shown in Table 1. As can be seen from Table 1, the longitudinal protection method proposed in this paper has high accuracy in photovoltaic-connected distribution network systems and strong applicability.
[0033] Table 1 Example 2 The present invention also provides a longitudinal protection device for a power distribution network, comprising: The first processing module is used to extract and repair key response data that is strongly related to line protection; The second processing module is used to construct longitudinal protection criteria based on Hellinger distance; The third processing module is used to implement longitudinal protection of the distribution network based on the key response data after abnormal data repair and the longitudinal protection criteria.
[0034] As one embodiment of the present invention, the first processing module uses cubic trigonometric Hermite interpolation with shape parameters to accurately repair abnormal current data.
[0035] As one embodiment of the present invention, the second processing module constructs a start-up criterion through an improved voltage gradient algorithm and calculates the probability distribution of the current at both ends of the faulty line using kernel density estimation, and then solves the Hellinger distance to construct a protection criterion; the third processing module combines data repair with Hellinger distance discrimination, and realizes fault discrimination and fault phase diagnosis within the area through abnormal data detection and repair, Hellinger distance calculation and threshold value comparison, and finally completes the longitudinal protection of the distribution network.
[0036] Example 3 The present invention also provides a distribution network longitudinal protection system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a distribution network longitudinal protection method when executed by the processor.
[0037] Example 4 The present invention also provides a storage medium storing a computer program, which executes a power distribution network longitudinal protection method when running.
[0038] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made to the technical solutions of the present invention by those skilled in the art without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.
Claims
1. A method for power distribution network pilot protection, characterized in that, include: Step 1: Extract and repair key response data that are strongly related to line protection; Step 2: Construct a longitudinal protection criterion based on Hellinger distance; Step 3: Implement longitudinal protection of the distribution network based on the key response data after abnormal data repair and the longitudinal protection criteria.
2. The power distribution network pilot protection method of claim 1, wherein, In step S1, the abnormal current data is accurately repaired using cubic trigonometric Hermite interpolation with shape parameters.
3. The power distribution network pilot protection method of claim 2, wherein, In step 2, an improved voltage gradient algorithm is used to construct a start-up criterion, and the probability distribution of the current at both ends of the faulty line is calculated using kernel density estimation. Then, the Hellinger distance is solved to construct a protection criterion. In step 3, data repair and Hellinger distance discrimination are combined. Through abnormal data detection and repair, Hellinger distance calculation and threshold value comparison, fault discrimination and fault phase diagnosis within the area are realized, and finally, the longitudinal protection of the distribution network is completed.
4. A power distribution network pilot protection device, characterized in that, include: The first processing module is used to extract and repair key response data that is strongly related to line protection; The second processing module is used to construct longitudinal protection criteria based on Hellinger distance; The third processing module is used to implement longitudinal protection of the distribution network based on the key response data after abnormal data repair and the longitudinal protection criteria.
5. The power distribution network pilot protection device of claim 4, wherein, The first processing module uses cubic trigonometric Hermite interpolation with shape parameters to accurately repair abnormal current data.
6. The distribution network longitudinal protection device as described in claim 5, characterized in that, The second processing module constructs a start-up criterion through an improved voltage gradient algorithm and uses kernel density estimation to calculate the probability distribution of the current at both ends of the faulty line, and then solves the Hellinger distance to construct a protection criterion. The third processing module combines data repair with Hellinger distance discrimination. Through abnormal data detection and repair, Hellinger distance calculation and threshold value comparison, it realizes fault discrimination and fault phase diagnosis within the area, and finally completes the longitudinal protection of the distribution network.
7. A longitudinal protection system for a power distribution network, characterized in that, include: A memory and a processor, wherein the memory stores a computer program executed by the processor, the computer program performing the distribution network longitudinal protection method as described in any one of claims 1-3 when executed by the processor.
8. A storage medium, characterized in that, The storage medium stores a computer program, which executes the power distribution network longitudinal protection method as described in any one of claims 1-3 when the computer program is running.