Intelligent analysis method for running state of power distribution network transformer
The method addresses inefficiencies and inaccuracies in transformer status analysis by using a sensor network and data-driven models to enhance real-time monitoring and fault detection in distribution networks.
Patent Information
- Application Number
- CN202411299096.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-18
- Publication Date
- 2024-12-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing transformer operation status analysis methods rely on subjective human judgment, lack systematicity and timeliness, leading to inefficient and inaccurate analysis results and untimely fault diagnosis, affecting the stability of distribution networks.
A method involving a sensor network, feature extraction, data analysis, and comprehensive prediction to analyze transformer status using sensors for electrical, environmental, and equipment parameters, with data preprocessing and modeling to determine abnormal conditions.
Enhances real-time monitoring and accuracy of transformer operation analysis, reducing fault occurrence by systematically analyzing transformer status through intelligent data processing.
Smart Images

Figure CN119093599A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of smart grids, and in particular to a method for intelligently analyzing variable operating states of distribution networks. Background Art
[0002] With the rapid development of my country's economy and the advancement of urbanization, energy demand has continued to grow. As a key link in energy transmission and distribution, the scale and structural complexity of the distribution network are also increasing year by year. The stable operation of the distribution network plays a vital role in ensuring the reliability, safety and economy of power supply. In this context, the transformer equipment in the distribution network, as an important hub connecting the upper power grid and end users, is directly related to the operating efficiency, power supply quality and user experience of the entire distribution network.
[0003] However, in the existing technology, the operation status analysis of the transformer often relies on the subjective experience and intuitive judgment of the operation and maintenance personnel. This method is not only inefficient, but also difficult to ensure the accuracy of the analysis results due to the influence of human factors. In addition, traditional analysis methods usually lack systematicity and timeliness, and it is difficult to cope with various complex situations that may arise in the operation of the distribution network. Sometimes it will lead to untimely diagnosis and handling of faults, which will affect the stable operation of the entire distribution network. Summary of the invention
[0004] The purpose of this section is to summarize some aspects of embodiments of the present invention and briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section and the specification abstract and the invention title of this application to avoid blurring the purpose of this section, the specification abstract and the invention title, and such simplifications or omissions cannot be used to limit the scope of the present invention.
[0005] In view of the problems existing in the above-mentioned existing transformer operation status analysis method, the present invention is proposed.
[0006] Therefore, the technical problem solved by the present invention is to solve the problems of low efficiency of the existing transformer operation status analysis method, difficulty in ensuring the accuracy of the analysis results, and untimely fault diagnosis and processing.
[0007] To solve the above technical problems, the present invention provides the following technical solutions: a method for intelligent analysis of the operating status of a distribution network transformer, comprising the following steps: a sensor network, which obtains status parameters through various types of sensors configured in the transformer system; a feature extraction module, which is wirelessly connected to the sensor network, and after obtaining the status parameters, adjusts and extracts parameters of different types to extract key feature parameters that can reflect the operating status of the transformer; a data analysis module, which is wirelessly connected to the feature extraction module, and after obtaining the key feature parameters, establishes data analysis models respectively; a comprehensive prediction module, which is data-connected to the data analysis module, obtains the status values output by each data analysis model, establishes a comprehensive prediction module, outputs a comprehensive prediction value, and determines whether the current transformer is abnormal based on the size of the comprehensive prediction value and a preset threshold; wherein the status parameters obtained by the sensor network specifically include: obtaining the transformer electrical current through the current sensor configured on the secondary side of the transformer system Parameters; obtain the transformer environment parameters through the temperature sensor, humidity sensor and pressure sensor configured in the transformer system; obtain the equipment parameters of the transformer equipment through the detector configured in the transformer system equipment; wherein, after obtaining the state parameters, it also includes data preprocessing of the parameter data; wherein the data preprocessing process specifically includes: cleaning, denoising and normalization; wherein, when the feature extraction module extracts the features of the transformer electrical parameters, it specifically includes the following steps: S1: within the rated time, the output terminals with a constant number of stable openings are continuously distributed in the transformer system; S2: obtain the current change curve, and synchronously select points with consistent time intervals within the rated time as reference points; S3: obtain the characteristic representative value in the current change curve as its key characteristic parameter, and the characteristic representative value specifically includes: maximum current value, minimum current value, average current value and current value at each reference point; wherein, the data analysis model established based on the transformer electrical parameters is specifically:
[0008] Among them, δ is the electrical state value of the transformer; I1 is the current value of the first reference point, A; I n is the current value of the nth reference point, A; n is the number of reference points; I min is the minimum current value, A; I max is the maximum current value, A; I 均 is the average current value, A; 1.302 and 1.64 are adjustment constants; dx is the integral operation; the data analysis model established based on the transformer environment parameters is as follows:
[0009] Among them, η is the environmental status value of the station; t is the temperature data of each point, ℃; n is the number of temperature detection points, pcs; t1 is the temperature data of the first temperature detection point, ℃; t nis the temperature data of the nth temperature detection point, °C; t min is the minimum value of the temperature data at each temperature detection point, ℃; t max is the maximum value of the temperature data of each temperature detection point, ℃; s is the humidity data of each point, %; m is the number of temperature detection points, pieces; s1 is the humidity data of the first humidity detection point, %; s m is the humidity data of the mth humidity detection point, %; s min It is the minimum value of humidity data at each humidity detection point, %; s max is the maximum value of the humidity data at each humidity detection point, %; 1.07, 1.84 and 0.11 are adjustment constants; dx is the integral operation; P is the pressure sensor data, Pa; Among them, the data analysis model established based on the equipment parameters of the station transformer is specifically:
[0010] Where φ is the equipment status value of the transformer; A is the insulation resistance value of the equipment, Ω; M is the abnormal temperature distribution area of the equipment, m 2 ; Wherein, the comprehensive prediction module constructed by the comprehensive prediction module is specifically:
[0011] Among them, μ is the comprehensive prediction value; δ is the electrical state value of the transformer; η is the environmental state value of the transformer; φ is the equipment state value of the transformer; t max is the maximum value of the temperature data at each temperature detection point, ℃; 1.86, 1.033 and 1.65 are adjustment constants; dx is the integral operation.
[0012] As a preferred solution of the intelligent analysis method for the variable operating status of the distribution network described in the present invention, the preset threshold is set to 2.16.
[0013] The present invention provides a method for intelligently analyzing the variable operation status of a distribution network, which has the following beneficial effects: 1. The present invention uses a sensor network to collect the operation data of the distribution network transformer in real time, thereby improving the real-time and accuracy of the monitoring data; 2. The present invention can accurately determine whether the operating status of the distribution network transformer is abnormal through feature extraction and data analysis, thereby reducing the probability of fault occurrence; The present invention solves the problems of low efficiency of existing transformer operation status analysis method, difficulty in ensuring the accuracy of analysis results, and untimely fault diagnosis and processing. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative labor. Among them: Figure 1 This is a flow chart of the overall method of the intelligent analysis method of the distribution network station variable operating status provided by the present invention.
[0015] Figure 2 This is a flow chart of a method for extracting features of transformer electrical parameters by the feature extraction module provided by the present invention. DETAILED DESCRIPTION
[0016] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.
[0017] In the existing technology, the operation status analysis of transformers often relies on the subjective experience and intuitive judgment of operation and maintenance personnel. This method is not only inefficient, but also difficult to ensure the accuracy of the analysis results due to the influence of human factors. In addition, traditional analysis methods usually lack systematicity and timeliness, and it is difficult to cope with various complex situations that may arise in the operation of the distribution network. Sometimes it will lead to untimely diagnosis and handling of faults, which will affect the stable operation of the entire distribution network.
[0018] Therefore, please refer to Figure 1 The present invention provides a method for intelligently analyzing the operation status of a distribution network station, comprising the following steps: Sensor network, which obtains status parameters through various types of sensors configured in the transformer system; The feature extraction module is connected to the sensor network by wireless data, and after obtaining the state parameters, it adjusts and extracts the parameters of different types to extract the key feature parameters that can reflect the operation state of the transformer; A data analysis module is wirelessly connected to the feature extraction module to obtain key feature parameters and establish data analysis models respectively; The comprehensive prediction module is connected to the data analysis module, obtains the status value output by each data analysis model, establishes a comprehensive prediction module, outputs a comprehensive prediction value, and determines whether the current transformer is abnormal based on the size of the comprehensive prediction value and a preset threshold.
[0019] Specifically, the state parameters acquired by the sensor network include: The electrical parameters of the transformer are obtained through the current sensor configured on the secondary side of the transformer system; The environmental parameters of the transformer are obtained through the temperature sensor, humidity sensor and pressure sensor configured in the transformer system; The device parameters of the station transformer device are obtained through the detector configured in the station transformer system device.
[0020] Specifically, the transformer electrical parameters are current parameter data obtained by the current sensor; Specifically, the station variable environmental parameters are temperature parameter data of each point obtained by the temperature sensor, humidity parameter data of each point obtained by the humidity sensor, and pressure parameter data obtained by the pressure sensor; Specifically, the equipment parameters of the transformer equipment are the insulation resistance value of the equipment and the abnormal temperature distribution area of the equipment.
[0021] Additional explanations are: 1. Measurement of insulation resistance: Measuring insulation resistance is usually done with an insulation resistance tester (also called a megohmmeter or high-voltage insulation tester). The measurement steps are as follows: Disconnect the power supply of the device under test to ensure safety.
[0022] Sufficiently discharge the device to avoid risk of electric shock during testing.
[0023] Connect the high voltage probe of the tester to the insulated part of the equipment.
[0024] Start the tester and gradually increase the test voltage, usually starting from a lower voltage and gradually increasing to the specified test voltage.
[0025] Measure and record the insulation resistance value displayed by the tester.
[0026] 2. Measurement of abnormal temperature distribution area of equipment: The temperature sensor is used for acquisition. The temperature sensor at each point can obtain the temperature range around the point, and the temperature range around all points is counted. When the temperature range near the selected point is 50% higher than the temperature range of any surrounding point, the point is an abnormal point. The measurement of abnormal temperature distribution area of equipment = number of abnormal points * temperature measurement area of each point.
[0027] It should be noted that current sensors, temperature sensors, humidity sensors, pressure sensors and detectors are all conventional sensors and will not be described in detail here.
[0028] Furthermore, after obtaining the state parameters, it also includes data preprocessing of the parameter data; The data preprocessing process specifically includes: cleaning, denoising and normalization. The data preprocessing steps are the application of existing conventional data processing steps, which are briefly described below: 1. Data cleaning: Identify and handle missing values: There may be missing data points in a dataset, which may be caused by equipment failure, data transmission errors, or other reasons. Methods for handling missing values include filling missing values (for example, using the mean, median, or mode), deleting records with missing values, or using interpolation methods (such as linear interpolation or spline interpolation) to estimate missing values.
[0029] Identify and handle outliers: Outliers may be caused by measurement errors, equipment failures, or abnormal operations. Methods for handling outliers include identifying outliers using box plots (IQR) or other statistical methods and considering removing or replacing these values.
[0030] Data transformation: For non-numeric data, data transformation, such as encoding or normalization, may be required to facilitate subsequent analysis and modeling.
[0031] 2. Data denoising: Time series smoothing: Use techniques such as moving average, low-pass filter, wavelet filter, etc. to smooth time series data to reduce the impact of noise.
[0032] Frequency domain filtering: The time series data is converted to the frequency domain through Fourier transform, and then the frequency domain filter is used to remove the unnecessary frequency components, and then it is returned to the time domain through inverse transform.
[0033] Kalman Filter: The Kalman filter is an efficient recursive filter that is able to estimate the state of a linear dynamic system in the presence of noise.
[0034] 3. Data normalization: Linear normalization: Scale the data to a fixed range, such as [0, 1] or [-1, 1].
[0035] Logarithmic Normalization: For log-transformed data, log normalization can reduce the proportional differences in the data.
[0036] Standardization: Standardization is the process of scaling the data to a normal distribution with a mean of 0 and a standard deviation of 1.
[0037] Furthermore, when the feature extraction module extracts features from the transformer electrical parameters, the steps specifically include: S1: Within the rated time, a constant number of output terminals are stably opened, and the transformer system continuously distributes power; S2: Obtain the current variation curve and simultaneously select points with consistent time intervals within the rated time as reference points; S3: Obtain characteristic representative values in the current variation curve as its key characteristic parameters, the characteristic representative values specifically include: maximum current value, minimum current value, average current value and current value at each reference point.
[0038] It should be noted that the rated time can be selected as 24 hours. When selecting points with consistent time intervals as reference points within the rated time, corresponding reference points can be selected every hour. The selection of the rated time and reference points does not affect the essence of the plan.
[0039] Furthermore, the data analysis model established based on the transformer electrical parameters is as follows:
[0040] Among them, δ is the electrical state value of the transformer; I1 is the current value of the first reference point, A; I n is the current value of the nth reference point, A; n is the number of reference points; I min is the minimum current value, A; I max is the maximum current value, A; I 均 is the average current value, A; 1.302 and 1.64 are adjustment constants; dx is the integral operation.
[0041] Furthermore, the data analysis model established based on the station variable environmental parameters is as follows:
[0042] Among them, η is the environmental status value of the station; t is the temperature data of each point, ℃; n is the number of temperature detection points, pcs; t1 is the temperature data of the first temperature detection point, ℃; t n is the temperature data of the nth temperature detection point, °C; t min is the minimum value of the temperature data at each temperature detection point, ℃; t max is the maximum value of the temperature data of each temperature detection point, ℃; s is the humidity data of each point, %; m is the number of temperature detection points, pieces; s1 is the humidity data of the first humidity detection point, %; s m is the humidity data of the mth humidity detection point, %; s min It is the minimum value of humidity data at each humidity detection point, %; s max is the maximum value of the humidity data at each humidity detection point, %; 1.07, 1.84 and 0.11 are adjustment constants; dx is the integral operation; P is the pressure sensing data, Pa.
[0043] Furthermore, the data analysis model established based on the equipment parameters of the transformer is as follows:
[0044] Where φ is the equipment status value of the transformer; A is the insulation resistance value of the equipment, Ω; M is the abnormal temperature distribution area of the equipment, m 2 .
[0045] Furthermore, the comprehensive prediction module constructed by the comprehensive prediction module is specifically:
[0046] Among them, μ is the comprehensive prediction value; δ is the electrical state value of the transformer; η is the environmental state value of the transformer; φ is the equipment state value of the transformer; t max is the maximum value of the temperature data at each temperature detection point, ℃; 1.86, 1.033 and 1.65 are adjustment constants; dx is the integral operation.
[0047] Specifically, the preset threshold is set to 2.16.
[0048] The present invention provides a method for intelligently analyzing the variable operation status of a distribution network, which has the following beneficial effects: 1. The present invention uses a sensor network to collect the operation data of the distribution network transformer in real time, thereby improving the real-time and accuracy of the monitoring data; 2. The present invention can accurately determine whether the operating status of the distribution network transformer is abnormal through feature extraction and data analysis, thereby reducing the probability of fault occurrence; The present invention solves the problems of low efficiency of existing transformer operation status analysis method, difficulty in ensuring the accuracy of analysis results, and untimely fault diagnosis and processing.
[0049] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should be included in the scope of the claims of the present invention.
Claims
1. A method for intelligent analysis of the variable operation status of a distribution network, characterized in that: The steps include: Sensor network, which obtains status parameters through various types of sensors configured in the transformer system; The feature extraction module is connected to the sensor network by wireless data, and after obtaining the state parameters, it adjusts and extracts the parameters of different types to extract the key feature parameters that can reflect the operation state of the transformer; A data analysis module is wirelessly connected to the feature extraction module to obtain key feature parameters and establish data analysis models respectively; A comprehensive prediction module is connected with the data analysis module to obtain the state value output by each data analysis model, establish a comprehensive prediction module, output a comprehensive prediction value, and determine whether the current transformer is abnormal based on the difference between the comprehensive prediction value and a preset threshold value; The state parameters acquired by the sensor network specifically include: The electrical parameters of the transformer are obtained through the current sensor configured on the secondary side of the transformer system; The environmental parameters of the transformer are obtained through the temperature sensor, humidity sensor and pressure sensor configured in the transformer system; Obtain equipment parameters of the transformer equipment through the detector configured in the transformer system equipment; Among them, after obtaining the state parameters, it also includes data preprocessing of the parameter data; Among them, the data preprocessing process specifically includes: cleaning, denoising and normalization; The feature extraction module specifically includes the following steps when extracting features from the transformer electrical parameters: S1: Within the rated time, a constant number of output terminals are stably opened, and the transformer system continuously distributes power; S2: Obtain the current variation curve and simultaneously select points with consistent time intervals within the rated time as reference points; S3: Acquire characteristic representative values in the current variation curve as key characteristic parameters, wherein the characteristic representative values specifically include: maximum current value, minimum current value, average current value and current value at each reference point; Among them, the data analysis model established based on the transformer electrical parameters is as follows: Among them, δ is the electrical state value of the transformer; I1 is the current value of the first reference point, A; I n is the current value of the nth reference point, A; n is the number of reference points; I min is the minimum current value, A; I max is the maximum current value, A; I 均 is the average current value, A; 1.302 and 1.64 are adjustment constants; dx is the integral operation; Among them, the data analysis model established based on the station variable environmental parameters is specifically: Among them, η is the environmental status value of the station; t is the temperature data of each point, ℃; n is the number of temperature detection points, pcs; t1 is the temperature data of the first temperature detection point, ℃; t n is the temperature data of the nth temperature detection point, °C; t min is the minimum value of the temperature data at each temperature detection point, ℃; t max is the maximum value of the temperature data of each temperature detection point, ℃; s is the humidity data of each point, %; m is the number of temperature detection points, pieces; s1 is the humidity data of the first humidity detection point, %; s m is the humidity data of the mth humidity detection point, %; s min It is the minimum value of humidity data at each humidity detection point, %; s max is the maximum value of the humidity data at each humidity detection point, %; 1.07, 1.84 and 0.11 are adjustment constants; dx is the integral operation; P is the pressure sensor data, Pa; Among them, the data analysis model established based on the equipment parameters of the transformer is as follows: Where φ is the equipment status value of the transformer; A is the insulation resistance value of the equipment, Ω; M is the abnormal temperature distribution area of the equipment, m 2 ; The comprehensive prediction module constructed by the comprehensive prediction module is specifically: Among them, μ is the comprehensive prediction value; δ is the electrical state value of the transformer; η is the environmental state value of the transformer; φ is the equipment state value of the transformer; t max is the maximum value of the temperature data at each temperature detection point, ℃; 1.86, 1.033 and 1.65 are adjustment constants; dx is the integral operation.
2. The method for intelligent analysis of the operation status of a distribution network station according to claim 1 is characterized in that: The preset threshold is set to 2.16.
Citation Information
Patent Citations
Factory affair equipment health management system based on Internet of Things data interconnection
CN118938837A
Multi-stage protection technology of intelligent integrated power supply system
CN119093589A
Cited By
Factory affair equipment health management system based on Internet of Things data interconnection
CN118938837A
