Method and system for life prediction of three-phase distribution transformer based on winding ratio

Abstract

The application provides a three-phase distribution transformer life prediction method and system based on winding ratio in the field of distribution transformer service life prediction, and aims to economically, efficiently and non-destructively predict the service life of a three-phase distribution transformer. The system comprises a communication module, an acquisition module, a storage module and a data processing module. The winding ratio is calculated according to the monitored voltage and current data of the three-phase distribution transformer. The sensitivity of the winding ratio to time is calculated according to the winding ratio on the current day and the winding ratio on the n-1 days before the current day. The life prediction model is established according to the sensitivity of the winding ratio to time, the winding ratio value under serious defects and the winding ratio on the current day. The value of the winding ratio on the current day is substituted into the life prediction model to calculate the remaining life since the current day.

Description

Technical Field

[0001] This invention relates to the field of power distribution transformer lifespan prediction, specifically to a method and system for predicting the lifespan of a three-phase power distribution transformer based on winding ratio. Background Technology

[0002] Transformers are crucial equipment in power systems, responsible for voltage level conversion and load distribution. Residential and industrial loads are widely distributed and large in capacity, requiring power supply through distribution transformers; therefore, power systems have a large number of distribution transformers. Currently, distribution transformers have a long service life, reaching over 50 years. However, due to rapid urban infrastructure development, the types of loads and total social load capacity are increasing exponentially. Meanwhile, the planned capacity of power systems in some areas is insufficient. These two factors lead to a large number of distribution transformers operating under heavy loads for extended periods. Under high load conditions, the likelihood of transformer defects increases significantly, and their service life decreases accordingly. To ensure high-quality electricity supply for users, real-time prediction of distribution transformer lifespan is necessary to guarantee their normal operation.

[0003] Existing methods for predicting the lifespan of distribution transformers are mainly divided into indirect and direct methods. The indirect method, namely the stress analysis method, is based on analytically determining the stress and strength data of the component materials. A computer then uses finite element analysis to calculate the degree of damage to the component and calculate the remaining lifespan accordingly. While it offers high accuracy, it requires corresponding stress detection equipment and computer simulation capabilities, resulting in high technical costs. The direct method typically employs destructive testing methods to obtain identical or similar samples. Destructive testing is then used to obtain the necessary data, including accelerated creep rupture tests and fatigue tests, to estimate the degree of lifespan damage and make a lifespan assessment. Although the difference between the direct and direct methods is small, the destructive testing damages transformer components, resulting in economic and time losses. Summary of the Invention

[0004] The purpose of this invention is to solve the problems existing in the prior art and provide a method for predicting the life of a three-phase distribution transformer based on the winding ratio, so as to solve how to predict the life of a three-phase distribution transformer economically, efficiently and without damage.

[0005] This invention is achieved through the following technical solution: a method for predicting the lifespan of a three-phase distribution transformer based on winding ratio, comprising the following steps:

[0006] The winding ratio is calculated based on the voltage and current data of the three-phase distribution transformer obtained from monitoring.

[0007] The sensitivity of the winding ratio to time is calculated based on the winding ratio of the current day and the winding ratio of the n-1 days prior to the current day.

[0008] A life prediction model is established based on the time sensitivity of the winding ratio, the winding ratio value under severe defects, and the winding ratio on the day.

[0009] The remaining life from that day is calculated by substituting the value of the winding ratio on that day into the life prediction model.

[0010] Furthermore, the general formula for calculating the winding ratio is as follows:

[0011]

[0012] Where, N i V represents the winding ratio of the i-th three-phase distribution transformer. i-1 V i-2 I represents the effective value of the primary side voltage and the effective value of the secondary side voltage of the three-phase distribution transformer, respectively. i Z i These represent the effective value of the secondary current and the secondary impedance value of the three-phase distribution transformer, respectively.

[0013] Furthermore, the average value of the winding ratio at different times of day (morning, noon, and evening) is calculated as the daily winding ratio.

[0014] Furthermore, the average value of the winding ratio is calculated by taking the winding ratio from 6 to 8 a.m., the winding ratio from 12 to 2 p.m., and the winding ratio from 6 to 20 p.m. as the daily winding ratio.

[0015] Furthermore, the general formula for calculating the time sensitivity of the winding ratio is as follows:

[0016]

[0017] Where, N iave(t) This represents the winding ratio on day t out of n days, including the current day, and N iave(n) Indicates the winding ratio for that day; This represents the average value of the winding ratio data over n days.

[0018] Furthermore, the lifetime prediction model is as follows:

[0019]

[0020] Where α1 represents the time sensitivity of the winding ratio; N f N represents the winding ratio under severe defects. iave(n) This indicates the winding ratio for that day.

[0021] Furthermore, the winding ratios of the current day and the n-1 days prior are used to form the original dataset. The noise in the original dataset is removed by a wavelet denoising algorithm to obtain a denoised dataset. The sensitivity of the winding ratio to time is calculated based on the denoised dataset, and the remaining lifespan is calculated based on the winding ratio of the current day in the denoised dataset.

[0022] Furthermore, the voltage and current data of the three-phase distribution transformer are obtained through smart meter monitoring.

[0023] This invention also provides a three-phase distribution transformer life prediction system based on winding ratio, used to implement the three-phase distribution transformer life prediction method based on winding ratio as described in this invention, and includes:

[0024] Communication module, acquisition module, storage module, data processing module;

[0025] The communication module is used to communicate with the power distribution network information management system in real time to obtain the voltage and current data recorded by the smart meters connected to the secondary side of each three-phase power distribution transformer in the power distribution network.

[0026] The acquisition module is used to collect voltage and current data recorded by smart meters connected to the secondary side of each three-phase distribution transformer at different times in the morning, noon and evening, and transmit them to the storage module.

[0027] The storage module is used to store the voltage and current datasets and remaining life information of each three-phase distribution transformer;

[0028] The data processing module is used to calculate the winding ratio, the time sensitivity of the winding ratio, and the remaining life, and sends the remaining life through the communication module.

[0029] Compared with the prior art, the beneficial effects of the present invention are:

[0030] 1. This invention can predict the remaining life of a three-phase distribution transformer by calculating the winding ratio based on the collected voltage and current data. If a smart meter for a three-phase distribution transformer is used, the voltage and current data can be obtained without the need to purchase additional stress detection equipment, thereby reducing hardware detection costs and improving the economic efficiency of power system operation.

[0031] 2. This invention only requires the detection of voltage and current operating data of a three-phase distribution transformer, avoiding damage to transformer components caused by destructive testing methods, and saving time and costs.

[0032] 3. This invention can be installed in a power distribution network monitoring system. The output information on the remaining lifespan of the three-phase power distribution transformer can be viewed by monitoring personnel in real time, which promotes the automation of the lifespan detection of the three-phase power distribution transformer, reduces the workload of personnel, and improves energy efficiency. Attached Figure Description

[0033] Figure 1 This is a flowchart of the three-phase distribution transformer life prediction method based on winding ratio in this specific embodiment;

[0034] Figure 2 This is the equivalent circuit diagram of the three-phase distribution transformer in this specific embodiment;

[0035] Figure 3 This is a structural diagram of the three-phase distribution transformer life prediction system based on winding ratio in this specific embodiment. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to the accompanying drawings:

[0037] This specific implementation method is illustrated using the winding ratio of a three-phase distribution transformer over a 30-day period as an example.

[0038] refer to Figure 1 As shown, a three-phase distribution transformer for which life prediction is required is selected. Based on the electrical connection relationship between the high and low voltage sides, the expression for calculating its winding ratio is written, and a linear regression prediction model for the winding ratio of the three-phase distribution transformer is established. In principle, the slope of the linear regression prediction model for the winding ratio of the three-phase distribution transformer reflects the sensitivity of the winding ratio to time. Therefore, the formula for calculating the slope is the formula for calculating the sensitivity of the winding ratio to time.

[0039] Under normal network conditions, the original three-phase voltage and current data of the three-phase distribution transformer are measured and collected through smart meters connected to the transformer. The turns ratio values ​​corresponding to the measurement data at 6:00, 12:00, and 18:00 on the same day are calculated, and the average winding turns ratio value of the three-phase distribution transformer on that day is calculated. A raw dataset of the three-phase distribution transformer winding turns ratios over 30 days is generated. The dataset is processed using a wavelet denoising algorithm to remove noise and improve the accuracy of the calculation. Based on the updated raw dataset of the three-phase distribution transformer winding turns ratios for the day, the parameter values ​​of the linear regression prediction model for the winding turns ratio are calculated. A severe defect value for the winding turns ratio is set. Based on the linear regression prediction model for the winding turns ratio with the determined parameters, a prediction model for the service life of the three-phase distribution transformer is established to calculate the remaining service life of the three-phase distribution transformer. Monitoring personnel can arrange the maintenance work of the distribution transformer based on the calculation results.

[0040] The equivalent circuit of a three-phase distribution transformer is based on Figure 2 As shown, the general formula for calculating the winding ratio based on the equivalent circuit is as follows:

[0041]

[0042] Where, N iV represents the winding ratio of the i-th three-phase distribution transformer. i-1 V i-2 I represents the effective value of the primary side voltage and the effective value of the secondary side voltage of the three-phase distribution transformer, respectively. i Z i These represent the effective value of the secondary current and the secondary impedance value of the three-phase distribution transformer, respectively.

[0043] Since the winding ratio of a three-phase distribution transformer changes slowly, it can be considered to change linearly with time. Therefore, its expression can be established as follows:

[0044]

[0045] Where α0 represents the linear regression constant, α1 represents the linear regression coefficient, and ε represents the random error.

[0046] The actual winding ratio values ​​of the three-phase distribution transformer are used to form the original dataset for the linear regression prediction model of the winding ratio, and its calculation expression is as follows:

[0047]

[0048] Where, N i(x) V i-1(x) V i-2(x) I i(x) These represent the winding ratio, effective values ​​of primary and secondary voltages, and effective value of secondary current of the three-phase distribution transformer at time x (x = 6, 12, 18), respectively.

[0049] The average winding ratio of the three-phase distribution transformer on that day is used to calculate the average level of the winding ratio of the three-phase distribution transformer on that day by combining measurement data at three different times. The calculation expression is as follows:

[0050]

[0051] Where, N iave This represents the average winding ratio of the three-phase distribution transformer on that day.

[0052] The wavelet denoising algorithm described above is used to remove the influence of abnormal signals such as large deviations in peak values ​​and data fluctuations in the original data of three-phase distribution transformer winding ratios within 30 days, thus removing noise and preventing noise from reducing calculation accuracy. Its principle is as follows:

[0053]

[0054] Where ψ a,b (t) represents the wavelet function, a represents the scaling factor, b represents the translation factor, * represents the conjugate complex number, f(x) is the input value, and W f(a,b) represents the wavelet transform signal of the input signal. The noise in the original data of the three-phase distribution transformer winding ratio within 30 days can be removed by wavelet transform and inverse wavelet transform to obtain the noise-reduced data set.

[0055] The formula for calculating the parameter values ​​of the linear regression prediction model for the winding ratio is:

[0056]

[0057] Where, N iave(t) This represents the winding ratio on day t out of 30 days, including the current day, and N iave(n) Indicates the winding ratio for the day; This represents the average winding ratio data over 30 days. The formula for its calculation is:

[0058]

[0059] The three-phase distribution transformer service life prediction model is used to predict and calculate the remaining service life of the three-phase distribution transformer through winding ratio. Its expression is as follows:

[0060]

[0061] Where α1 represents the time sensitivity of the winding ratio; N f The winding ratio under severe defects is determined empirically; N iave(n) This indicates the winding ratio for that day.

[0062] refer to Figure 3 As shown in the figure, this specific embodiment also provides a three-phase distribution transformer life prediction system based on winding ratio, used to implement the three-phase distribution transformer life prediction method based on winding ratio in this specific embodiment, and includes:

[0063] Communication module, acquisition module, storage module, data processing module.

[0064] The communication module is used to communicate with the power distribution network information management system in real time to obtain the voltage and current data recorded by the smart meters connected to the secondary side of each three-phase power distribution transformer in the power distribution network.

[0065] The acquisition module is used to collect voltage and current data recorded by smart meters connected to the secondary side of each three-phase distribution transformer at different times in the morning, noon and evening, and transmit them to the storage module.

[0066] The storage module is used to store the voltage and current datasets and remaining life information of each three-phase distribution transformer;

[0067] The data processing module is used to calculate the winding ratio, the time sensitivity of the winding ratio, and the remaining life, and sends the remaining life through the communication module.

[0068] The above technical solution is only one embodiment of the present invention. For those skilled in the art, based on the principles disclosed in the present invention, it is easy to make various types of improvements or modifications, and not limited to the technical solutions described in the specific embodiments of the present invention. For example, when the accuracy requirement is not high, the sensitivity of the winding ratio to time and the life prediction model can also be calculated based on the original dataset. Therefore, the above description is only a preferred option and does not have a limiting meaning.

Claims

1. A method for predicting the lifespan of a three-phase distribution transformer based on winding ratio, characterized in that, Includes the following steps: The winding ratio is calculated based on the voltage and current data of the three-phase distribution transformer obtained from monitoring. According to the day and the days prior The calculation of the winding ratio over time determines the sensitivity of the winding ratio to time. A life prediction model is established based on the time sensitivity of the winding ratio, the winding ratio value under severe defects, and the winding ratio on the day. Substitute the value of the winding ratio on that day into the life prediction model to calculate the remaining life from that day onwards; The general formula for calculating the time sensitivity of the winding ratio is as follows: in, N iave(t) Represents the nth day within the n-day period, including the current day. t The winding ratio of the day, and Indicates the winding ratio for the day; This represents the average value of the winding ratio data over n days. The lifespan prediction model is as follows: in, This indicates the sensitivity of the winding ratio to time. N f This represents the winding ratio under severe defects. This indicates the winding ratio for that day.

2. The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio according to claim 1, characterized in that, The general formula for calculating the winding ratio is as follows: in, N i Representing the The winding ratio of a three-phase distribution transformer. V i-1 , V i-2 These represent the effective values ​​of the primary voltage and the effective value of the secondary voltage of the three-phase distribution transformer, respectively. I i , Z i These represent the effective value of the secondary current and the secondary impedance value of the three-phase distribution transformer, respectively.

3. The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio according to claim 1, characterized in that, The average value of the winding ratio at different times of day (morning, noon, and evening) is used as the daily winding ratio.

4. The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio according to claim 3, characterized in that, The average value of the winding ratio is calculated as the daily winding ratio, taking the winding ratios from 6 to 8 a.m., 12 to 2 p.m., and 6 to 8 p.m.

5. The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio according to claim 1, characterized in that, On that day and before that day The original dataset is composed of the winding ratios of each day. The noise in the original dataset is removed by a wavelet denoising algorithm to obtain a denoised dataset. The sensitivity of the winding ratios to time is calculated based on the denoised dataset, and the remaining lifespan is calculated based on the winding ratios of the day in the denoised dataset.

6. The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio according to claim 1, characterized in that, Voltage and current data of a three-phase distribution transformer obtained through smart meter monitoring.

7. A three-phase distribution transformer life prediction system based on winding ratio, characterized in that, The method for predicting the lifespan of a three-phase distribution transformer based on winding ratio as described in claim 3, and includes: Communication module, acquisition module, storage module, data processing module; The communication module is used to communicate with the power distribution network information management system in real time to obtain the voltage and current data recorded by the smart meters connected to the secondary side of each three-phase power distribution transformer in the power distribution network. The acquisition module is used to collect voltage and current data recorded by smart meters connected to the secondary side of each three-phase distribution transformer at different times in the morning, noon and evening, and transmit them to the storage module. The storage module is used to store the voltage and current datasets and remaining life information of each three-phase distribution transformer; The data processing module is used to calculate the winding ratio, the time sensitivity of the winding ratio, and the remaining life, and sends the remaining life through the communication module.

Images