A transformer fault detection and positioning method based on voltage analysis

Abstract

The application provides a transformer fault detection positioning method based on voltage analysis, which comprises the following steps: step (1), collecting three-phase voltage signals of the transformer respectively; step (2), performing abnormal data processing on the collected three-phase voltage signals of the transformer; step (3), performing cluster analysis on the three-phase voltage signals of the transformer; and step (4), judging whether the transformer has an abnormality and a specific fault phase. The application provides a transformer fault detection positioning method based on voltage analysis, which can accurately judge whether the transformer has a fault and determine a fault phase.

Description

Technical Field

[0001] This invention belongs to the field of power detection technology, and specifically relates to a transformer fault detection and location method based on voltage analysis. Background Technology

[0002] Power transformers are fundamental equipment for power transmission and distribution, widely used in industry, agriculture, transportation, and urban communities. Transformer malfunctions may not always present obvious symptoms, requiring effective monitoring for timely detection. Their stable and reliable operation is crucial for ensuring the safety and reliability of the power grid and electrical equipment.

[0003] This invention proposes a transformer fault detection and location method based on voltage analysis. It can determine the specific faulty phase by collecting three-phase voltage signals, performing cluster analysis, and finally comparing the cluster centers obtained from the cluster analysis with the reference values ​​of each phase. Summary of the Invention

[0004] This invention provides a transformer fault detection and location method based on voltage analysis, which can accurately determine whether a transformer has a fault and identify the faulty phase.

[0005] Specifically, this invention relates to a transformer fault detection and location method based on voltage analysis, the transformer fault detection and location method comprising the following steps:

[0006] Step (1): Collect the three-phase voltage signals of the transformer respectively;

[0007] Step (2): Perform abnormal data processing on the collected three-phase voltage signals of the transformer;

[0008] Step (3): Perform cluster analysis on the three-phase voltage signals of the transformer;

[0009] Step (4): Determine whether the transformer has any abnormalities and specific fault phases.

[0010] The method for abnormal data processing of the collected three-phase voltage signals of the transformer is as follows:

[0011] First, calculate the average voltage signal of each phase. U At U is the instantaneous value of phase A voltage. Bt U is the instantaneous value of phase B voltage. Ct Here, is the instantaneous value of phase C voltage, and T is the sampling period;

[0012] Secondly, calculate the standard deviation of phase A voltage. Standard deviation of phase B voltage C-phase voltage standard deviation

[0013] Next, determine whether it is satisfied. ε is set to 1.2; if so, the bad data is repaired: Where α+β=1;

[0014] Determine if it satisfies ε is set to 1.2; if so, the bad data is repaired: Where α+β=1;

[0015] Determine if it satisfies ε is set to 1.2; if so, the bad data is repaired: Where α+β=1.

[0016] The method for clustering analysis of the three-phase voltage signals of the transformer includes:

[0017] (1) Calculate the phase voltage imbalance rate PVUR of the transformer;

[0018] (2) Extract the three-phase voltage signals corresponding to the 30 moments when the PVUR of each phase is the largest, and construct a voltage sequence U of length 30. A (n), U B (n), U C (n), n = 1, 2, ..., 30;

[0019] (3) Use the rated three-phase voltage of the transformer as a sample;

[0020] (4) Set the number of clusters K=3, and use the average value of the three-phase voltage signal of the transformer as the initial cluster center;

[0021] (5) Calculate the voltage sequence U A (n), U B (n), U C (n) Euclidean distance from each cluster center;

[0022] (6) Assign the voltage sequence to the cluster center with the smallest Euclidean distance;

[0023] (7) Calculate the mean of the samples in each class and use it as the new cluster center;

[0024] (8) If the cluster center and the samples of each class no longer change, or the number of iterations reaches a predetermined value, end; otherwise, return to (5).

[0025] The algorithm for calculating the phase voltage imbalance rate of the transformer is as follows:

[0026] First, calculate the average value of the three-phase voltage signal of the transformer. U A U is the effective value of phase A voltage. BU is the effective value of phase B voltage. C This is the effective value of the C-phase voltage;

[0027] Secondly, calculate the phase voltage imbalance rate of the transformer.

[0028] Calculate the voltage sequence U A (n), U B (n), U C The algorithm for calculating the Euclidean distance between (n) and each cluster center is as follows:

[0029] Where A i Corresponding to the voltage sequence U A (n), U B (n), U C (n), B i Corresponding to the cluster center.

[0030] The specific algorithm for determining whether the transformer has any abnormalities and specific fault phases is as follows:

[0031] In step (3), the cluster center obtained at the end is compared with the reference value of each phase. If it is greater than the reference value, the transformer is faulty and the corresponding phase is the faulty phase.

[0032] Compared with existing technologies, the beneficial effects are: the transformer fault detection and location method mainly collects and processes abnormal data of each phase voltage, then performs cluster analysis, and finally compares the cluster center obtained from the cluster analysis with the reference value of each phase to determine the specific fault phase. Attached Figure Description

[0033] Figure 1 This is a flowchart illustrating the process of a transformer fault detection and location method based on voltage analysis according to the present invention. Detailed Implementation

[0034] The following detailed description, with reference to the accompanying drawings, of a transformer fault detection and location method based on voltage analysis according to the present invention.

[0035] like Figure 1 As shown, the transformer fault detection and location method of the present invention includes the following steps:

[0036] Step (1): Collect the three-phase voltage signals of the transformer respectively;

[0037] Step (2): Perform abnormal data processing on the collected three-phase voltage signals of the transformer;

[0038] Step (3): Perform cluster analysis on the three-phase voltage signals of the transformer;

[0039] Step (4): Determine whether the transformer has any abnormalities and specific fault phases.

[0040] The method for abnormal data processing of the collected three-phase voltage signals of the transformer is as follows:

[0041] First, calculate the average voltage signal of each phase. U At U is the instantaneous value of phase A voltage. Bt U is the instantaneous value of phase B voltage. Ct Here, is the instantaneous value of phase C voltage, and T is the sampling period;

[0042] Secondly, calculate the standard deviation of phase A voltage. Standard deviation of phase B voltage C-phase voltage standard deviation

[0043] Next, determine whether it is satisfied. ε is set to 1.2; if so, the bad data is repaired: Where α+β=1;

[0044] Determine if it satisfies ε is set to 1.2; if so, the bad data is repaired: Where α+β=1;

[0045] Determine if it satisfies ε is set to 1.2; if so, the bad data is repaired: Where α+β=1.

[0046] The method for clustering analysis of the three-phase voltage signals of the transformer includes:

[0047] (1) Calculate the phase voltage imbalance rate of the transformer. in U A U is the effective value of phase A voltage. B U is the effective value of phase B voltage. C This is the effective value of the C-phase voltage;

[0048] (2) Extract the three-phase voltage signals corresponding to the 30 moments when the PVUR of each phase is the largest, and construct a voltage sequence U of length 30. A (n), U B (n), U C (n), n = 1, 2, ..., 30;

[0049] (3) Use the rated three-phase voltage of the transformer as a sample;

[0050] (4) Set the number of clusters K=3, and use the average value of the three-phase voltage signal of the transformer as the initial cluster center;

[0051] (5) Calculate the voltage sequence U A (n), U B (n), U C (n) Euclidean distance from each cluster center;

[0052] (6) Assign the voltage sequence to the cluster center with the smallest Euclidean distance;

[0053] (7) Calculate the mean of the samples (i.e., voltage sequences) for each class and use it as the new cluster center;

[0054] (8) If the cluster center and the samples of each class no longer change, or the number of iterations reaches a predetermined value, end; otherwise, return to (5).

[0055] Calculate the voltage sequence U A (n), U B (n), U C The algorithm for calculating the Euclidean distance between (n) and each cluster center is as follows:

[0056] Where A i Corresponding to the voltage sequence U A (n), U B (n), U C (n), B i Corresponding to the cluster center.

[0057] Finally, it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention and not intended to limit it. Those skilled in the art should understand that modifications or equivalent substitutions can be made to the specific embodiments of the present invention, but such modifications or alterations are all within the scope of protection of the pending claims.

Claims

1. A transformer fault detection and location method based on voltage analysis, characterized in that, The transformer fault detection and location method includes the following steps: Step (1): Collect the three-phase voltage signals of the transformer respectively; Step (2): Perform abnormal data processing on the collected three-phase voltage signals of the transformer; Step (3): Perform cluster analysis on the three-phase voltage signals of the transformer, specifically including: (31) Calculate the phase voltage imbalance rate PVUR of the transformer; (32) Extract the three-phase voltage signals corresponding to the 30 times when the PVUR of each phase is the largest, and construct a voltage sequence U of length 30. A (n), U B (n), U C (n), n = 1, 2, ..., 30; (33) The rated three-phase voltage of the transformer is used as a sample; (34) Set the number of clusters K=3, and use the average value of the three-phase voltage signal of the transformer as the initial cluster center; (35) Calculate the voltage sequence U A (n), U B (n), U C (n) is the Euclidean distance from each cluster center; (36) Assign the voltage sequence to the cluster center with the smallest Euclidean distance; (37) Calculate the mean of the samples in each class and use it as the new cluster center; (38) If the cluster centers and samples of each class no longer change, or the number of iterations reaches a predetermined value, end; otherwise, return to (35); Step (4): Determine whether the transformer has any abnormalities and specific fault phases.

2. The transformer fault detection and location method based on voltage analysis according to claim 1, characterized in that, The method for abnormal data processing of the collected three-phase voltage signals of the transformer is as follows: First, calculate the average voltage signal of each phase. U At U is the instantaneous value of phase A voltage. Bt U is the instantaneous value of phase B voltage. Ct Here, is the instantaneous value of phase C voltage, and T is the sampling period; Secondly, calculate the standard deviation of phase A voltage. Standard deviation of phase B voltage C-phase voltage standard deviation Next, determine whether it is satisfied. ε is set to 1.2; if so, the abnormal data is repaired: Where α+β=1; Determine if it satisfies ε is set to 1.2; if so, the abnormal data is repaired: Where α+β=1; Determine if it satisfies ε is set to 1.2; if so, the abnormal data is repaired: Where α+β=1.

3. The transformer fault detection and location method based on voltage analysis according to claim 1, characterized in that, The algorithm for calculating the phase voltage imbalance rate of the transformer is as follows: First, calculate the average value of the three-phase voltage signal of the transformer. U A U is the effective value of phase A voltage. B U is the effective value of phase B voltage. C This is the effective value of the C-phase voltage; Secondly, calculate the phase voltage imbalance rate of the transformer.

4. The transformer fault detection and location method based on voltage analysis according to claim 3, characterized in that, Calculate the voltage sequence U A (n), U B (n), U C The algorithm for Euclidean distance between (n) and each cluster center is as follows: Where A i Corresponding to the voltage sequence U A (n), U B (n), U C (n), B i Corresponding to the cluster center, i is the number of voltage signal sampling points.

5. The transformer fault detection and location method based on voltage analysis according to claim 4, characterized in that, The specific algorithm for determining whether the transformer has any abnormalities and specific fault phases is as follows: In step (3), the cluster center obtained at the end is compared with the reference value of each phase. If it is greater than the reference value, the transformer is faulty and the corresponding phase is the faulty phase.

Images