CN-122021010-A - Mutual inductor reliability prediction method and system based on ocean environmental factor coupling field

CN122021010ACN 122021010 ACN122021010 ACN 122021010ACN-122021010-A

Abstract

The method comprises the steps of obtaining relevant data of marine environmental factors and the type of a transformer material, constructing the marine environmental factor coupling field based on the relevant data, wherein the marine environmental factor coupling field comprises a salt spray transmission coupling field, a wet heat circulation coupling field and a mechanical stress coupling field, establishing a performance degradation mapping mechanism between the marine environmental factor coupling field and the type of the transformer material, obtaining a performance degradation track of the transformer by combining the performance degradation mapping mechanism and a state estimation algorithm, and obtaining a reliability estimation result of the transformer based on the performance degradation track of the transformer.

Inventors

HU XIAOXU
LIN YUXIN
HUANG TIANFU
ZHANG YING
WU ZHIWU
LIN TONGYAO
JIANG XINYU
WANG WENJING
CHEN ZILIN
Tong Chengxin

Assignees

国网福建省电力有限公司营销服务中心
国网福建省电力有限公司

Dates

Publication Date: 20260512
Application Date: 20260126

Claims (10)

1. The method for predicting the reliability of the mutual inductor based on the marine environment factor coupling field is characterized by comprising the following steps of: Acquiring relevant data of marine environmental factors and the type of mutual inductor materials; Constructing a marine environmental factor coupling field based on the related data, wherein the marine environmental factor coupling field comprises a salt spray transmission coupling field, a damp-heat circulation coupling field and a mechanical stress coupling field; Establishing a performance degradation mapping mechanism between the marine environment factor coupling field and the type of the mutual inductor material, and combining the performance degradation mapping mechanism and a state estimation algorithm to obtain a performance degradation track of the mutual inductor; And obtaining a reliability evaluation result of the transformer based on the performance degradation track of the transformer.
2. The method for predicting the reliability of the mutual inductor based on the marine environmental factor coupling field according to claim 1, wherein the salt spray transmission coupling field is calculated in the following manner: ; In the formula, Representing the position; Representing the time; representing the salt fog-wind speed influence coefficient; Representation of At the time position Is set at the wind speed of (2); Representation of The salt source intensity at the moment; representing the salt fog-humidity influence coefficient; Representation of Time of day at position Is a relative humidity of (2); Representation of Spatial gradient of the salt source intensity at the moment; Representing the salt spray-temperature influence coefficient; Representation of Time of day at position Is set at a temperature of (2); Representation of Time of day at position Is a salt concentration of (1); The calculation mode of the damp-heat circulation coupling field is as follows: ; In the formula, Representing the temperature-humidity influence coefficient; representing the temperature-air pressure influence coefficient; Representing the temperature-radiation influence coefficient; Representation of Time of day at position Is a gas pressure of (2); Representation of Time of day at position Is used for the solar radiation intensity of (a); the mechanical stress coupling field is calculated in the following way: ; In the formula, Representing the combined action coefficient of wind and waves; Representing the wind turbulence influence coefficient; representing the hot air coupling influence coefficient; Representation of Time of day at position Wave height of (2); Representation of Time of day at position Is a turbulent intensity of (2); Representation of Time of day at position Is a temperature change rate of (a).
3. The method for predicting reliability of a transformer based on a coupling field of marine environmental factors according to claim 1, wherein a performance degradation mapping mechanism between the coupling field of marine environmental factors and the type of material of the transformer is established, expressed as: ; , ; In the formula, Representation of Performance of the instant transformer; Representing initial transformer performance; Represent the first The comprehensive degradation rate caused by the coupling field of the marine environmental factors; Represent the first Marine environment factor coupling field Time of day at position Strength of (2); representing a time index; Representing a preset first A seed marine environmental factor coupling field effect index; Representing a preset first Performance autocorrelation index of the seed marine environmental factor coupling field; Represent the first Temperature dependent weighting of the type of instrument transformer material; Represent the first The type of the material of the seed mutual inductor is in the first A marine environmental factor couples a material sensitivity factor of the field.
4. The method for predicting the reliability of the transformer based on the marine environmental factor coupling field according to claim 3, wherein the performance degradation track of the transformer is obtained by combining a performance degradation mapping mechanism and a state estimation algorithm, and specifically comprises the following steps: establishing a degraded track state vector; inputting the state vector of the degradation track into a state transition equation to obtain a predicted state vector at the next moment; Carrying out self-adaptive tracking on the current state vector by using a Kalman filter to obtain the observed noise covariance at the current moment; Calculating a Kalman filter coefficient based on the observed noise covariance at the current moment and a preset observation matrix; correcting the predicted state vector at the next moment based on the Kalman filter coefficient to obtain the state vector at the next moment; Iterating the process until the state vectors at all moments are obtained, and splicing the state vectors at all moments to obtain the performance degradation track of the transformer.
5. The method for predicting the reliability of the mutual inductor based on the marine environmental factor coupling field according to claim 1, wherein the reliability evaluation result of the mutual inductor is obtained based on the performance degradation track of the mutual inductor, and the specific steps are as follows: the method comprises the steps of calculating the total strength of the marine environment factor coupling field based on the strength of each marine environment factor coupling field, identifying marine environment modes based on the total strength of the marine environment factor coupling field, and establishing a corresponding reliability model for each marine environment mode; reliability prediction is carried out on the current marine environment mode based on the reliability model, and a reliability score is obtained, namely a reliability evaluation result of the transformer is obtained; And mapping the reliability score based on a multi-level early warning mechanism to obtain and execute an early warning decision.
6. The method for predicting reliability of a transformer based on a coupling field of marine environmental factors according to claim 1, further comprising monitoring a current marine environmental pattern and calculating a sequential probability ratio of the current marine environmental pattern; Judging whether the sequential probability ratio of the current marine environment mode exceeds an environment failure threshold value, if so, triggering marine environment mode switching, and switching to the corresponding marine environment mode according to the sequential probability ratio.
7. The method for predicting reliability of a transformer based on a coupling field of marine environmental factors according to claim 6, wherein the sequential probability ratio of the current marine environmental pattern is calculated by: ; In the formula, Representation of Sequential probability ratio of time instants; Indicating that under the original assumption, observe Probability density of (c); Representation of The coupling field strength of the marine environmental factors at moment; Representing alternative hypotheses; representing the original hypothesis.
8. The method for predicting the reliability of the transformer based on the marine environmental factor coupling field according to claim 6, wherein the environmental failure threshold is calculated by the following steps: ; In the formula, Representation of A time of day environmental failure threshold; Representing an initial threshold; Represent the first The environmental factor sensitivity coefficient of the marine environmental factor coupling field; Representation of Time of day (time) Coupling field strength by marine environmental factors; Representation of Time of day (time) Maximum intensity of the seed marine environmental factor coupling field; Represent the first An index value of the marine environmental factor coupling field.
9. The method for predicting the reliability of a transformer based on a coupling field of marine environmental factors according to claim 6, wherein the environmental failure threshold is optimized by using a gradient descent method, and the method is expressed as: ; ; In the formula, Representing a preset learning rate; Representing the gradient; an optimization objective function representing an environmental failure threshold; Represent the first A plurality of observation data points; the number of observation points is represented; representing a preset false alarm rate weight; representing preset report missing rate weight; representing a preset maintenance cost weight; Represent the first False positive rates of individual observation data points; Represent the first The rate of missing report of the individual observed data points; Represent the first Maintenance costs for individual observation data points.
10. Mutual inductor reliability prediction system based on marine environment factor coupling field, characterized in that the system comprises: The data acquisition module acquires relevant data of the marine environmental factors and the type of the mutual inductor material; the coupling field construction module is used for constructing a marine environmental factor coupling field based on the related data, wherein the marine environmental factor coupling field comprises a salt spray transmission coupling field, a wet heat circulation coupling field and a mechanical stress coupling field; the track prediction module is used for establishing a performance degradation mapping mechanism between the marine environment factor coupling field and the type of the transformer material, and combining the performance degradation mapping mechanism and a state estimation algorithm to obtain a performance degradation track of the transformer; and the evaluation module is used for obtaining a reliability evaluation result of the transformer based on the performance degradation track of the transformer.

Description

Mutual inductor reliability prediction method and system based on ocean environmental factor coupling field Technical Field The invention relates to the technical field of intelligent operation and maintenance of offshore wind power equipment, in particular to a method and a system for predicting reliability of a transformer based on a marine environmental factor coupling field. Background With the transformation of global energy structures to clean energy, offshore wind power is an important renewable energy form, and is developed on a large scale. The high-voltage transformer is used as core equipment in a booster station and a power transmission line of the offshore wind farm, plays key roles of voltage/current measurement, signal transmission and relay protection, and the operation reliability of the high-voltage transformer is directly related to the safe and stable operation and power supply continuity of the whole wind farm. However, the offshore environment has extremely strong complexity and corrosiveness, and the long-term synergistic effect of the salt fog, high humidity, severe temperature and humidity circulation, strong wind load, wave impact and other multi-environmental factors can lead to the ageing of materials of the mutual inductor, structural damage and performance degradation speed far faster than that of land equipment. According to statistics, the failure occurrence rate of the offshore wind power transformer is 3-5 times that of land similar equipment, the failure maintenance difficulty is high, the cost is high, and the economic benefit and the power supply reliability of a wind power plant are seriously affected. Therefore, accurate reliability prediction and early warning of the offshore wind power transformer are realized, and the method becomes a core requirement in the field of intelligent operation and maintenance of offshore wind power equipment. The current technical means for reliability prediction of the transformer is mainly developed around the traditional equipment fault diagnosis and reliability assessment method, and specifically comprises the following steps of a fault diagnosis method based on electrical parameters, a multi-element environmental factor linear analysis method, an acceleration test method based on single stress and a traditional statistics and physical modeling method; although the prior art has certain application in reliability evaluation under land equipment or simple environment, the method has fundamental defects aiming at complex operation scenes of the marine wind power transformer, mainly comprises the following aspects of firstly, limiting environmental modeling, modeling by adopting a linear superposition mode by taking factors such as salt fog, temperature, humidity and wind load in the marine environment as independent variables, completely neglecting complex coupling field effect in the marine environment, failing to truly reflect comprehensive action mechanisms of the environment on the transformer, secondly, insufficient cognition of degradation mechanisms, incapability of deeply exploring cooperative degradation mechanisms between the marine environment coupling field and the material, structure and performance of the transformer by the prior method, causing lack of physical basis of a model, insufficient scientificity and reliability of a prediction result, thirdly, self-adaptive capacity deficiency that environmental dynamic changes are difficult to be dealt with, marine environment has strong randomness, nonlinearity and mode mutation characteristics, traditional model parameters are fixed, prediction accuracy is easy to be reduced by self-adaptive adjustment prediction strategies according to real-time changes of the environment coupling field, fourthly, prediction accuracy and timeliness are insufficient, the prior method is extremely low in cost of prediction accuracy and actual maintenance, and the prior art is difficult to be deployed in advance, and the prior art is difficult to reduce the actual maintenance requirements of the wind power equipment is low in the aspects of prediction accuracy and is easy to be maintained The combination of false alarm and false omission results in lack of economic consideration for early warning decision, easy occurrence of excessive maintenance or untimely maintenance, and failure to maximize operation and maintenance benefits. Therefore, there is a need to establish a method for predicting a transformer with environmental adaptation capability, which can reflect the inherent physical association between the environment and the device. Disclosure of Invention In order to solve the problems in the prior art, the invention provides a mutual inductor reliability prediction method and system based on a marine environment factor coupling field. The technical scheme of the invention is as follows: in one aspect, the invention provides a method for predicting reliability of a mutual inductor based