CN-121978582-A - Air-core reactor turn-to-turn short circuit fault diagnosis and positioning method based on magnetic field sensor array

CN121978582ACN 121978582 ACN121978582 ACN 121978582ACN-121978582-A

Abstract

The invention relates to a diagnosis and positioning method of an inter-turn short circuit fault of an air core reactor based on a magnetic field sensor array, which belongs to the field of detection of electric equipment and comprises the following steps of arranging the magnetic field sensor array at the outer axial position of the air core reactor to obtain magnetic field distribution data near the air core reactor, calculating the magnetic field distribution data of the position of the magnetic field sensor array of the air core reactor under normal working conditions and inter-turn short circuit fault working conditions at different positions, constructing a complete fault characteristic database, setting inter-turn short circuit faults at the determined position of the air core reactor, carrying out data calibration based on measured data and data of the fault characteristic database, calculating the measured data of the air core reactor through a fault diagnosis algorithm, determining whether the air core reactor has faults or not, and accurately positioning the fault position. The invention can diagnose turn-to-turn short circuit faults, realize accurate positioning of fault points and improve the operation safety and reliability of power equipment.

Inventors

HEI GUOYU
XU ZHENG

Assignees

重庆大学

Dates

Publication Date: 20260505
Application Date: 20260402

Claims (8)

1. The air-core reactor turn-to-turn short circuit fault diagnosis and positioning method based on the magnetic field sensor array is characterized by comprising the following steps of: arranging a magnetic field sensor array at the axial position outside the air core reactor to acquire magnetic field distribution data near the air core reactor; calculating magnetic field distribution data of the magnetic field sensor array position of the air core reactor under the normal working condition and the turn-to-turn short circuit fault working condition at different positions, and constructing a complete fault characteristic database; Step three, setting turn-to-turn short circuit faults at the determined positions of the air core reactors, and carrying out data calibration based on measured data and data of a fault characteristic database; step four, calculating actual measurement data of the air reactor through a fault diagnosis algorithm to determine whether the reactor has faults and accurately position fault positions, wherein the fault diagnosis algorithm in the step four comprises three stages: taking magnetic field phase data acquired by a magnetic field sensor arranged at the axial center of the air core reactor as a reference, and making a difference value with magnetic field phase data acquired by other magnetic field sensors; The second stage, carrying out fitness function calculation on the phase difference value of the calibrated actually measured magnetic field and the phase difference value of the database under the normal working condition, if the convergence condition is met, the reactor works normally, and if the convergence condition is not met, the reactor has turn-to-turn short circuit fault; And thirdly, performing fitness function calculation on the phase difference value of the calibrated actually measured magnetic field and the phase difference value of the database through a genetic algorithm based on a population competition mechanism, and outputting the currently calculated fault position result if convergence conditions are met.
2. The method for diagnosing and positioning the turn-to-turn short circuit fault of the air-core reactor based on the magnetic field sensor array, which is disclosed in claim 1, is characterized in that the magnetic field sensor array is composed of nine TMR magnetic field sensors, the sensors are uniformly and equidistantly arranged from the upper end to the lower end of the air-core reactor, and the sensitive direction of the TMR magnetic field sensors is the axial direction.
3. The method for diagnosing and positioning the turn-to-turn short circuit fault of the air reactor based on the magnetic field sensor array according to claim 2, wherein the magnetic field distribution data are phase data extracted from magnetic field time domain signals acquired by the TMR magnetic field sensor, and the phase data extraction method is to obtain the phase by performing sine fitting on the magnetic field signals through a least square method.
4. The diagnosis and positioning method for turn-to-turn short circuit faults of air reactor based on magnetic field sensor array as claimed in claim 3, wherein in the second step, a complete fault characteristic database is constructed by interpolation mode, comprising the following steps: Let x i denote that the ith envelope fails, y denote the position of the failure point in the axial direction of the reactor, and for two adjacent failure positions Q 1 (x i ,y 1 ) and Q 2 (x i ,y 2 at envelope x i ), the magnetic field data for any failure position Q (x i , y) therebetween is calculated by: Wherein f (Q 1 ) and f (Q 2 ) respectively represent the magnetic field distribution data of the fault positions Q 1 (x i ,y 1 ) and Q 2 (x i ,y 2 ), and f (Q) is the magnetic field distribution result obtained by linear interpolation calculation, and in this way, the magnetic field distribution data corresponding to all the fault positions are obtained, so that the construction of the fault characteristic database is completed.
5. The method for diagnosing and locating an inter-turn short circuit fault of an air-core reactor based on a magnetic field sensor array as set forth in claim 4, wherein the step three of performing data calibration based on the measured data and the data of the fault characteristic database comprises the following steps: With measured data With fault signature database data There is the following linear relationship: And (5) performing linear regression fitting by a least square method to obtain parameters a and b.
6. The method for diagnosing and positioning the turn-to-turn short circuit fault of the air reactor based on the magnetic field sensor array as set forth in claim 5, wherein the specific calculation process in the first stage is as follows: Each TMR magnetic field sensor in the magnetic field sensor array is numbered 1-9 in sequence from the lower end to the upper end, the TMR magnetic field sensor at the axial center of the air core reactor is numbered 5, and the phase difference value of the kth sensor meets the following formula: Obtaining the measured magnetic field phase difference value through calculation Magnetic field phase difference value under normal working condition And fault signature database magnetic field phase difference value 。
7. The method for diagnosing and positioning the turn-to-turn short circuit fault of the air reactor based on the magnetic field sensor array as set forth in claim 6, wherein the second stage comprises the following specific processes: Measured magnetic field phase difference value And the magnetic field phase difference value under normal working condition Is a fitness function of (a) Satisfies the following formula: setting convergence discrimination parameters If (3) Indicating that the air-core reactor is working normally if And indicating that the air-core reactor has turn-to-turn short circuit fault.
8. The method for diagnosing and positioning the turn-to-turn short circuit fault of the air reactor based on the magnetic field sensor array as set forth in claim 7, wherein the third stage comprises the following specific calculation process: Firstly, determining the envelope where the fault occurs, and then determining the axial position where the specific fault of the turn-to-turn short circuit of the air reactor is located based on the envelope position, setting the envelope number as i, setting each envelope as an initial population 1~i independently, setting the population scale as 100, and adapting the fitness function Satisfies the following formula: The i populations are independently subjected to iterative computation for 3 times through a genetic algorithm, and the optimal fitness function of each population is obtained: corresponding axial position The i population optimal fitness functions compete with each other to select an optimal population x, and the following formula is satisfied: x is the package number of the turn-to-turn short circuit fault; Under the premise that the x-envelope fails, adopting a genetic algorithm to continue iterative optimization calculation and setting convergence discrimination parameters If (3) And the algorithm converges, and the calculated and output position y is the axial position of the final fault occurrence, so that the fault positioning is completed.

Description

Air-core reactor turn-to-turn short circuit fault diagnosis and positioning method based on magnetic field sensor array Technical Field The invention belongs to the technical field of power equipment detection, and relates to a diagnosis and positioning method for turn-to-turn short circuit faults of an air core reactor based on a magnetic field sensor array. Background The air-core reactor is used as key high-voltage primary equipment in a power system, plays an important role in reactive power compensation, system voltage regulation and stabilization, short-circuit current limitation and the like, and the running state of the air-core reactor is directly related to the safety and reliability of a power grid. Therefore, the method for guaranteeing the safe and stable operation of the air core reactor is an important link for maintaining the stability of the power system. The air-core reactor is usually installed in open air for a long time, and is easily influenced by factors such as temperature and humidity change, pollution erosion, system overvoltage and the like. Under the action, the winding may be subjected to phenomena such as damp, partial discharge, partial overheating and the like, so that the inter-turn insulation performance is reduced or even damaged, and finally inter-turn short circuit is induced. The turn-to-turn short circuit is one of the most common fault types in the operation process of the air-core reactor, and has prominent harmfulness, if the fault equipment cannot be found and cut off in time, the current distribution near the fault point is abnormal, the local temperature rise of the coil is obvious, the insulation aging is further accelerated, the fault is gradually expanded from a small quantity of turn-to-turn short circuits to a plurality of turn-to-turn short circuits, and the reactor burnout and even firing accidents can be caused when the fault is serious, so that the safety and the stability of the power system are obviously threatened. Therefore, the development of accurate detection and early recognition of turn-to-turn short circuit faults has important engineering significance and application value. The existing diagnosis of the turn-to-turn short circuit of the air-core reactor mainly depends on multi-source signals including electric signals, thermal signals, mechanical signals, vibration signals and the like. The method based on the electric parameters generally judges faults through characteristic changes such as voltage, current, power angle or equivalent impedance, and researches are also carried out to fuse the electric signals with environmental information such as temperature, humidity, wind speed and the like so as to improve the reliability of diagnosis. Meanwhile, turn-to-turn shorts change electromagnetic force distribution and cause structural vibration characteristics to be abnormal, and related vibration characteristics have been used for fault identification and localization in combination with machine learning. However, in general, the method has the defects of low sensitivity to early local faults, easiness in being influenced by environmental noise and fluctuation of operation conditions and limited accurate positioning capability on inter-turn short circuits of the reactor. From a mechanistic point of view, a larger circulating current will flow in the shorted turns, around which a significantly enhanced local magnetic field will be generated according to the biot-savart law. Therefore, the fault detection by using the magnetic field information around the reactor has a clear physical basis. Compared with the traditional means such as electricity, heat, vibration and the like, the magnetic field measurement has the advantages of non-contact, high sensitivity, strong real-time performance and the like, and provides a more potential technical path for early identification and positioning of turn-to-turn short circuits. Disclosure of Invention In view of the above, the present invention aims to provide a method for diagnosing and positioning turn-to-turn short circuit faults of air core reactors based on a magnetic field sensor array. In order to achieve the above purpose, the present invention provides the following technical solutions: A diagnosis and positioning method for turn-to-turn short circuit faults of an air reactor based on a magnetic field sensor array comprises the following steps: arranging a magnetic field sensor array at the axial position outside the air core reactor to acquire magnetic field distribution data near the air core reactor; calculating magnetic field distribution data of the magnetic field sensor array position of the air core reactor under the normal working condition and the turn-to-turn short circuit fault working condition at different positions, and constructing a complete fault characteristic database; Step three, setting turn-to-turn short circuit faults at the determined positions of the air core reactors, and ca