CN101556299A - Method for detecting micro deformation of transformer winding - Google Patents

Abstract

The invention discloses a method for detecting the small deformation of a transformer winding, which includes: inputting a pulse signal with a frequency higher than 1KHz into the transformer, collecting the signals of the input terminal and the output terminal of the transformer through a sensor, processing the collected signals, and calculating the output terminal and the output terminal. For the transfer function at the input end, the correlation coefficient is calculated through the calculated transfer function and the transfer function before the transformer is used for the first time. If it is judged that the correlation coefficient is less than the threshold, an alarm signal for winding deformation will be issued. If it is judged that the correlation coefficient is greater than or equal to the threshold, then The processed data of the collected signals are stored in the database. The method of the invention has simple measurement steps and is easy to implement, and the threshold value of the correlation coefficient can be adjusted to realize the measurement accuracy adjustment of the transformer deformation, which overcomes the long-standing technical difficulty in transformer deformation detection and is easy to popularize.

Description

Method for Detecting Small Deformation of Transformer Winding

technical field

The invention relates to detection technology, in particular to a method for detecting small deformation of transformer windings.

Background technique

Among the various equipment in the power system, the transformer is one of the more expensive and important equipment, and its safety performance is of great significance to ensure the safety of the power grid. Statistics on transformers over the years have shown that 110kV and above power transformers in the country are damaged due to external short-circuit faults, accounting for 50% of the total number of accidents. Judging from the dismantling inspection, most of them are caused by winding deformation (such as axial and radial dimension changes, body displacement, winding distortion, bulging, etc.), and the main reason for these deformations is that the transformer insulation has been running for a long time. The electrothermal aging in the process leads to performance degradation or even damage, and the resulting electromotive force and mechanical force act on the winding during external short circuit. The four main influencing factors of transformer insulation aging are mechanical force, heat, electric field and oil pollution. Transformer outlet short-circuit current and the electromotive force generated by closing inrush are the main mechanical factors affecting the aging of transformer insulation. Strong short-circuit current can cause winding deformation and damage to oil-paper insulation. Smaller short-circuit current or closing inrush electromotive force will cause The small deformation and gradually accelerates this process, and the resulting local defects of oil-paper insulation will generate partial discharge in the electric field, further accelerating the deterioration of local insulation; thermal aging is the main form of transformer oil-paper insulation aging, which causes slow chemical degradation of oil-paper insulation. degradation, resulting in a decline in insulation performance. At present, the maximum allowable hot spot temperature of power transformers can reach 140°C, which has a significant impact on the insulation performance of oil-paper.

Traditional transformer test methods include: transformation ratio measurement, winding resistance, short-circuit impedance and loss, excitation impedance, dielectric loss, etc. Other special tests include: partial discharge, frequency response, vibration analysis, water in oil, infrared detection, recovery voltage and degree of polymerization of insulating paper, etc.

The conventional test method is used to detect the overall performance index of the transformer; in the special test, the frequency response method and the vibration analysis method can detect the deformation of the winding. The former has a better application effect and has been widely used in the power industry, but it cannot detect the small deformation of the winding. . The above-mentioned transformer deformation detection can detect large deformations, but cannot detect small deformations of windings.

Contents of the invention

In view of this, the present invention provides a method for detecting the small deformation of the transformer winding, so as to solve the above-mentioned problem that the detection of the small deformation of the winding cannot be detected for the transformer deformation.

In order to solve the above problems, the present invention provides a method for detecting small deformations of transformer windings, including:

The pulse signal with a frequency higher than 1KHZ is used to input the transformer, and the signals of the input and output terminals of the transformer are collected through the sensor. After the collected signals are processed, the transfer function of the output and input terminals is calculated, and the calculated transfer function is compared with the transformer for the first time. The correlation coefficient is calculated by using the transfer function before use, and if it is judged that the correlation coefficient is smaller than the threshold value, an alarm signal of winding deformation is issued.

Preferably, said processing the collected signal includes:

The acquired signal is denoised by wavelet windowing, and the denoised signal is subjected to time-frequency transformation to obtain the frequency function.

Preferably, the threshold is 0.6.

Preferably, the method further includes: if it is judged that the correlation coefficient is greater than or equal to the threshold value, then storing the processed data of the collected signals in the database.

The method of the invention has simple measurement steps and is easy to implement, and the threshold value of the correlation coefficient can be adjusted to realize the measurement accuracy adjustment of the transformer deformation, which overcomes the long-standing technical difficulty in transformer deformation detection and is easy to popularize.

Description of drawings

Figure 1 is an equivalent network diagram of a transformer winding;

Figure 2 is a schematic diagram of equipment connection during detection;

Fig. 3 is a flowchart of the detection method.

Detailed ways

In order to clearly illustrate the solutions in the present invention, preferred embodiments are given below and detailed descriptions are given in conjunction with the accompanying drawings.

Firstly, the principle of the method of the present invention is set forth. For a transformer, when the frequency is high (over 1kHz), the transformer iron core basically does not work, and each winding can be regarded as a coil consisting of distributed parameters such as resistance, capacitance, and inductance. passive linear two-port network. The equivalent passive linear two-port network can be seen in Figure 1. Set the distributed inductance per unit length of the winding, the longitudinal capacitance and the capacitance to ground as L, K, and C respectively, and ignore the resistance of the winding (usually very small). The effective network can analyze its characteristics through the transfer function H(jω). If the winding has mechanical deformation such as axial and radial dimension changes, it will inevitably change the distribution parameters of the network such as L, K, and C, resulting in changes in the zero and pole distribution of its transfer function H(jω). Through this feature, it is possible to detect changes in the distribution of zeros and poles of the transfer function H(jω) when the transformer is not in use, and to detect changes in the distribution of zeros and poles of the transfer function H(jω) again during the regular inspection after the transformer is used. . By comparing the H(jω) obtained by the two tests, it is possible to judge whether the state of the transformer winding is deformed. In general, the frequency response analysis method is to measure the transfer function of the transformer winding in a wide frequency band, analyze the frequency response characteristics of the winding, and judge the state of the winding.

The small deformation of the transformer winding and the aging of the insulation lead to a small change in the network parameters. In order to ensure the accuracy of the measurement, the input pulse signal is required to have a higher amplitude and a steeper wave front (nanosecond level). It is difficult for general signal generators to meet the requirements at the same time, so the parameters of the pulse signal generator are shown in Table 1:

Output voltage ±0.1～2kV±5％

Pulse Polarity Positive/Negative

Pulse repetition frequency 0.1～999.9kHz±10％

Burst Length Standard 15ms 2.5kHz/5kHz

     Pulse Adjustable The number of pulses can be selected from 1 to 200

Pulse leading edge 5ns±30％

Pulse Width (at 50Ω) 50ns±30％

Pulse width (at 1KΩ) 35ns～150ns

Pulse train cycle 300ms

Internal resistance 50Ω±20%

Dimensions (mm) 440×430×180(D×W×H)

Machine weight ＜15kg

Power supply Single-phase AC220V±10%, 50/60Hz

Table 1

Set the pulse generator at one end of the transformer, as shown in Figure 2, the two voltage clamps of the transformer are respectively connected to the corresponding voltage sensors (voltage probes), the voltage reaches the voltage sensor after voltage division, and the voltage sensor that collects the high voltage signal measures the voltage It is 6kV, the input impedance is 50MΩ, and the bandwidth is 400MHz. In order to collect the voltage signal at the first and last ends of the input transformer winding and input it to the analysis terminal.

Since the rising edge of the pulse signal input to the transformer winding is very steep (5ns±30%), the most basic function realized by the analysis terminal is to collect this extremely steep rising edge pulse, and the sampling rate is greater than 2GS/S. The analysis terminal uses the wavelet windowing method to denoise the collected signals to remove some external interference signals. Then time-frequency transform is performed on the denoised signal, and the function of the frequency of the signal at the first and last ends of the transformer is obtained. The end is: V _o (f), and the function of the first end (input end) is: V _i (f). Calculate the transfer function, that is, the frequency response characteristic of the winding, the formula is: $h\left(f\right)=\frac{V_o\left(f\right)}{V_i\left(f\right)}.$ Compare the measured and calculated frequency response characteristic H(f) of the transformer winding with its frequency characteristic in a good state (initial measurement before the transformer leaves the factory or is put into operation), and calculate the correlation coefficient of the consistency between the two, and pass the correlation coefficient To judge whether the transformer has a small deformation of the winding, because different transformers have different electrical characteristics, it is preferable to select a correlation coefficient of less than 0.6, then it is considered that there is deformation, otherwise the test data will be stored in the database.

The testing process can be seen in Figure 3, including:

Step 1: Set the sampling frequency, the number of times used, and the related settings for saving the sampling data;

Step 2: Collect transformer data;

Step 3: Constantly judge the number of sampling times, if the predetermined number is not reached, go to step 1, otherwise go to step 4;

Step 4: Save adoption data;

Step 5: Perform wavelet denoising, Fourier transform, and spectrum analysis on the collected signal to calculate the transfer function;

Step 6: extracting the measured transfer function of the transformer before use;

Step 7: compare the data in step 5 and step 6, and calculate the correlation coefficient;

Step 8: Determine whether the correlation coefficient is less than the threshold, usually set the threshold to 0.6, and adjust the corresponding correlation coefficient if the measurement accuracy needs to be higher; if so, alarm; otherwise, store the data in step 5 into the database.

The method of the invention has simple measurement steps and is easy to implement, and the threshold value of the correlation coefficient can be adjusted to realize the measurement accuracy adjustment of the transformer deformation, which overcomes the long-standing technical difficulty in transformer deformation detection and is easy to popularize.

For the methods described in the various embodiments of the present invention, any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims (4)

1, detects the method for micro deformation of transformer winding, it is characterized in that, comprise: adopt frequency to be higher than the pulse signal input transformer of 1KHZ, by the input end of sensor acquisition transformer and the signal of output terminal, after the signal that collects handled, calculate the transport function of output terminal and input end, use preceding transport function to calculate related coefficient first by transport function and the transformer that calculates, less than threshold value, then send the alerting signal of winding deformation if judge related coefficient.

2, method according to claim 1 is characterized in that, the described signal that collects is handled comprises:

Adopt small echo to window the signal of gathering is carried out denoising, the signal after the denoising is carried out time-frequency conversion, obtain frequency function.

3, method according to claim 1 is characterized in that, described threshold value is 0.6.

4, method according to claim 1 is characterized in that, also comprises: if judge related coefficient more than or equal to threshold value, the data storage after then the described signal that collects being handled is to database.

Images