Rapid differential protection method for low-frequency transformer

By monitoring the transformer's flux saturation state and short data window differential current calculation technology, the problem of reduced sensitivity of traditional differential protection methods in low-frequency power transmission systems is solved, and the transformer's fast and high-reliability differential protection is achieved.

CN120109738APending Publication Date: 2025-06-06NANJING TUONA ELECTRIC CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510269329.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-03-07
Publication Date
2025-06-06

AI Technical Summary

Technical Problem

In low-frequency power transmission systems, the sensitivity of the traditional transformer differential protection method is reduced and is susceptible to interference, resulting in mislocking of protection or delayed action, affecting reliability.

Method used

By monitoring whether the transformer's magnetic flux enters the saturation zone, combined with the short data window differential current calculation technology, the flux value is calculated in real time and the magnetic saturation state is determined to achieve fast differential protection.

Benefits of technology

It improves the sensitivity and reliability of the differential protection of the transformer, quickly recognizes the excitation surge current, reduces misjudgment, and is suitable for low-frequency complex working conditions and new energy power electronic equipment.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure BDA0005302193830000021
    Figure BDA0005302193830000021
  • Figure BDA0005302193830000022
    Figure BDA0005302193830000022
  • Figure BDA0005302193830000023
    Figure BDA0005302193830000023
Patent Text Reader

Abstract

The invention discloses a rapid differential protection method for a low-frequency transformer, and the method comprises the steps: carrying out the recognition of magnetizing inrush current through judging whether the low-frequency transformer is subjected to magnetic saturation or not, recognizing the magnetizing inrush current when the magnetic flux of the transformer is greater than a saturation inflection point set value, and locking differential protection, and recognizing the fault when the magnetic flux of the transformer is smaller than the saturation inflection point, and opening the differential protection. And meanwhile, a short-time data window differential current amplitude calculation method is designed, and the differential protection action outlet time is not greater than 1 / 3 of low-frequency cyclic waves. According to the method, the magnetizing inrush current is identified by judging whether the low-frequency transformer enters the magnetic saturation state or not, and the first principle generated by the magnetizing inrush current of the transformer is utilized for identification, so that the method has relatively high applicability and is suitable for various operation and fault working conditions such as transformer air-drop, restorative inrush current, sympathetic inrush current and air-drop faults; the method is not influenced by high fault harmonic content in a transformer region of a low-frequency power transmission system, has high reliability and sensitivity, avoids the risk of misjudgment caused by the fact that a traditional magnetizing inrush current criterion is easily influenced by harmonic waves, is high in action speed, and realizes faster and more reliable protection of the low-frequency transformer. The method can be further popularized and applied to industrial frequency transformer differential protection in a wide application scene of new energy power electronic equipment.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention belongs to the technical field of relay protection for electric power systems, and specifically relates to a fast differential protection method for low-frequency transformers, which can be widely applied to transformer differential protection in application scenarios of new energy power electronic equipment. Background Art

[0002] At present, the traditional transformer differential protection method mainly identifies the excitation surge current and internal fault current through harmonic components (such as second harmonics). However, due to the influence of current limitation in low-frequency transmission systems, the short-circuit current of the same type of fault is much smaller than that of conventional systems, and the sensitivity of longitudinal differential protection is reduced. At the same time, the harmonic content of the differential current is high during faults in the transformer area, and the traditional harmonic judgment criteria are easily interfered with, which may cause the protection to be falsely locked or delayed, affecting the reliability of protection. In addition, the low-frequency cycle time is longer than the power frequency, and the fault removal is slow according to the traditional differential element full-cycle algorithm. Therefore, there is an urgent need for a fast and reliable differential protection method based on the essential principles of the electromagnetic characteristics of transformers. Summary of the invention

[0003] In order to make up for the shortcomings of traditional transformer differential principle in low-frequency application scenarios, the present invention proposes a fast differential protection method for low-frequency transformers. By monitoring whether the transformer flux enters the saturation region to identify the excitation surge current, combined with the short data window differential current calculation technology, fast and high-reliability differential protection for low-frequency transformers is achieved.

[0004] In order to achieve the above objectives, the present invention adopts the following technical solutions.

[0005] A low-frequency transformer fast differential protection method, characterized in that it comprises the following steps:

[0006] (1) Transformer voltage iterative integration calculation of transformer flux value

[0007] By collecting the voltage signals on both sides of the transformer, dynamic approximate fitting is performed in combination with the transformer flux attenuation characteristics and the preset attenuation time constant. The transformer flux value is calculated in real time using the iterative integration method. The specific formula is:

[0008]

[0009] Among them, u(n) is the voltage of the transformer at the current sampling point, φ(n) is the flux value calculated at the current sampling point, φ(n-1) is the flux value calculated at the previous sampling point, and N is the number of sampling points per cycle. τ is the transformer flux decay time constant, which is generally in seconds.

[0010] Δφ dc(n-1) is the attenuation value of the DC flux at the previous moment at the current moment, which is an approximate fit of the DC flux attenuation. Considering that the transformer flux attenuation time is as long as seconds, and the protection action time is in the order of tens of milliseconds, the use of a faster attenuation fitting has less impact on the protection logic judgment. Therefore, the flux attenuation change can be simply calculated by the following formula:

[0011]

[0012] T is the low frequency period, φ dc (n-1) is the DC component of the magnetic flux at the previous sampling point. φ dc (n-1) can be simply approximated by the following formula:

[0013]

[0014] (2) Magnetic saturation state determination

[0015] The saturation inflection point constant φset is set, and its value is determined according to the saturation inflection point of the transformer core material, generally 105% to 120% of the rated voltage flux.

[0016] Compare the magnetic flux value φ(n) with φset in real time.

[0017] When φ(n)≥φset, the transformer is judged to be in magnetic saturation state, the differential current is identified as magnetizing inrush current, and the differential protection is locked;

[0018] When φ(n)<φset, it is judged as a fault state and the differential protection is opened.

[0019] (3) Fast differential current calculation and action criteria

[0020] A short data window filter algorithm is used to calculate the differential current amplitude within a sampling data window that does not exceed 1 / 3 of the low-frequency cycle (for example, a low-frequency 20 Hz system corresponds to a time window ≤ 16.7 ms) after the fault occurs.

[0021] The selection of short data window filter is as follows,

[0022] Sine fundamental filter:

[0023] Cosine Fundamental Filter:

[0024] x(nk) is the data of M sampling points, b s (k) and b) c (k) is the filter coefficient, y s (n), y c (n) are the calculated real and imaginary parts of the fundamental wave respectively.

[0025] bs (k) is obtained by solving the following equation,

[0026] A*b c =v c ;

[0027] A*b s =v s ;

[0028] Where A is

[0029]

[0030] N is the number of sampling points per cycle, j is the harmonic to be filtered out, and when j is 0, the first line in the matrix is ​​for filtering out DC.

[0031] v s =[0,0,…,0,-1] T ;

[0032] v c =[0,0,…,1,0] T ;

[0033] The differential current amplitude can be obtained by calculating the real and imaginary parts of the differential current through the short data window filter. When the differential current amplitude exceeds the action setting value and the magnetic saturation state determines the open protection, the differential protection action outlet.

[0034] The beneficial effects of the present invention include:

[0035] (1) Identify the excitation inrush current based on the physical characteristics of magnetic saturation, avoid the misjudgment problem caused by the interference of fault harmonics in traditional harmonic judgment criteria, and improve the protection reliability;

[0036] (2) The short data window algorithm makes the protection action time ≤ 1 / 3 cycle, which helps to quickly remove the fault and improve the protection speed;

[0037] (3) It is suitable for various complex low-frequency fault conditions such as no-load closing, response inrush current, restorative inrush current and high harmonics. It can also be widely used in general transformer differential protection in the application scenarios of new energy power electronic equipment. DETAILED DESCRIPTION

[0038] A fast differential protection method for a low-frequency transformer comprises the following steps:

[0039] (1) Transformer voltage iterative integration to calculate transformer flux value

[0040] By collecting the voltage signals on both sides of the transformer, dynamic approximate fitting is performed in combination with the transformer flux attenuation characteristics and the preset attenuation time constant. The transformer flux value is calculated in real time using the iterative integration method. The specific formula is:

[0041]

[0042] Among them, u(n) is the voltage of the transformer at the current sampling point, φ(n) is the flux value calculated at the current sampling point, φ(n-1) is the flux value calculated at the previous sampling point, and N is the number of sampling points per cycle. τ is the transformer flux decay time constant, which is generally in seconds.

[0043] Δφ dc (n-1) is the attenuation value of the DC flux at the previous moment at the current moment, which is an approximate fit of the DC flux attenuation. Considering that the transformer flux attenuation time is as long as seconds, and the protection action time is in the order of tens of milliseconds, the use of a faster attenuation fitting has less impact on the protection logic judgment. Therefore, the flux attenuation change can be simply calculated by the following formula:

[0044]

[0045] T is the low frequency period, φ dc (n-1) is the DC component of the magnetic flux at the previous sampling point. φ dc (n-1) can be simply approximated by the following formula:

[0046]

[0047] For microcomputer protection, φ(n) is digitally calculated as the infinite integral of the voltage sampling value. After power-on, φ(n-1) and φ are calculated for the first time. dc The initial value of (n-1) is set to 0.

[0048] (2) Magnetic saturation state determination

[0049] The saturation inflection point constant φset is set, and its value is determined according to the saturation inflection point of the transformer core material, generally 105% to 120% of the rated voltage flux. According to the above flux value iterative calculation formula, the flux value has been uniformly converted to a voltage value, and φset is the transformer saturation inflection point voltage.

[0050] Compare the magnetic flux value φ(n) with φset in real time.

[0051] When φ(n)≥φset, the transformer is judged to be in magnetic saturation state, the differential current is identified as magnetizing inrush current, and the differential protection is locked;

[0052] When φ(n)<φset, it is judged as a fault state and the differential protection is opened.

[0053] Since the transformer will inevitably desaturate within one cycle, the magnetic saturation state needs to be returned with a delay.

[0054] (3) Fast differential current calculation and action criteria

[0055] A short data window filter algorithm is used to calculate the differential current amplitude within a sampling data window that does not exceed 1 / 3 of the low-frequency cycle (for example, a low-frequency 20 Hz system corresponds to a time window ≤ 16.7 ms) after the fault occurs.

[0056] The selection of short data window filter is as follows,

[0057] Sine fundamental filter:

[0058] Cosine Fundamental Filter:

[0059] x(nk) is the data of M sampling points, b s (k) and b c (k) is the filter coefficient, y s (n), y c (n) are the calculated real and imaginary parts of the fundamental wave respectively.

[0060] b s (k) is obtained by solving the following equation,

[0061] A*b c =v c ;

[0062] A*b s =v s ;

[0063] Taking 48 sampling points per cycle as an example, take a data window larger than 1 / 4 cycle, if M is 13, then A is

[0064]

[0065] v s =[0,0,…,0,-1] T ;

[0066] v c =[0,0,…,1,0] T ;

[0067] The differential current amplitude can be obtained by calculating the real and imaginary parts of the differential current through the short data window filter. When the differential current amplitude exceeds the action setting value and the magnetic saturation state determines the open protection, the differential protection action outlet.

[0068] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A fast differential protection method for a low-frequency transformer, characterized in that: The following steps are included: (1) Obtaining the transformer flux value by iterative integration of the transformer voltage; (2) When the magnetic flux value is greater than the set value of the saturation inflection point, it is identified as an excitation inrush current and the differential protection is locked. When the transformer magnetic flux value is less than the saturation inflection point, it is identified as a fault and the differential protection is opened; (3) The differential current is calculated by a fast short window filter algorithm. When the differential current amplitude is greater than a fixed value and the flux excitation surge current identification principle opens the differential protection, the differential action output is activated.

2. The transformer voltage iterative integration method for obtaining the transformer flux value according to claim 1 is characterized in that: Fitting is performed based on the transformer flux decay characteristics and decay time constant.

3. The fast short window filter according to claim 1, characterized in that: The differential current amplitude is calculated using the sampling values ​​within a short data window that does not exceed 1 / 3 cycle after the fault occurs.