Search

CN-121829675-B - Monitoring method and system for gas-shielded medium-voltage switch cabinet

CN121829675BCN 121829675 BCN121829675 BCN 121829675BCN-121829675-B

Abstract

The application relates to the technical field of switch cabinet monitoring, in particular to a monitoring method and a system of a gas protection medium-voltage switch cabinet. A collaborative monitoring system integrating pressure, temperature and current multidimensional information is constructed, and the limitation of single parameter monitoring is broken through. The Bayesian probability and the D-S evidence theory are innovatively combined, so that objectivity of statistical data is utilized, evidence conflict and cognition uncertainty are properly processed, and diagnosis is upgraded from threshold alarm to probabilistic trust evaluation. The application can effectively monitor the current state type of the switch cabinet in real time, drive the operation and maintenance mode to change from periodic overhaul to predictive maintenance, and greatly reduce the operation and maintenance cost and the unplanned power failure risk while ensuring the power supply reliability.

Inventors

  • DONG SHIJUN
  • ZHANG MIN
  • HE ZHE

Assignees

  • 成都新创界自动化设备有限公司

Dates

Publication Date
20260508
Application Date
20260313

Claims (10)

  1. 1. The monitoring method of the gas protection medium voltage switch cabinet is characterized by comprising the following steps: Acquiring an overall temperature sequence, an air pressure sequence, a busbar load sequence and an actual temperature rise sequence of a plurality of contact points, which are acquired in a target time period in the gas protection medium-voltage switch cabinet, wherein the target time period comprises a plurality of time points of target duration before a current time point; the target time period is divided into a plurality of time windows, the air pressure sequences of the time windows are compensated based on the whole temperature sequences of the time windows to obtain a standard air pressure sequence at the standard temperature, and the air pressure trend and the confidence coefficient of the air pressure trend are extracted from the standard air pressure sequence; calculating theoretical temperature rise of the contact point based on a pre-constructed contact point temperature rise prediction model and bus loads of a plurality of time windows, calculating residual errors of the theoretical temperature rise and the actual temperature rise of the contact point of the time windows, extracting a temperature rise abnormality index of the contact point corresponding to the maximum residual error of each time window, and calculating an actual temperature rise average value of the contact points corresponding to the current time window; the method comprises the steps of establishing an air pressure trend sequence based on air pressure trends of a plurality of time windows, establishing a temperature rise abnormality index sequence based on temperature rise abnormality indexes of the time windows, and calculating the correlation degree between the air pressure trend sequence and the temperature rise abnormality index sequence; And constructing a feature vector based on the air pressure trend of the current time window, the confidence coefficient of the air pressure trend of the current time window, the temperature rise abnormality index of the current time window, the actual temperature rise average value of the current time window and the correlation degree, and inputting the feature vector into a fusion model constructed in advance to obtain a monitoring result, wherein the fusion model fuses a Bayesian diagnosis model and a DS evidence theory.
  2. 2. The method for monitoring a gas-shielded medium-voltage switchgear according to claim 1, wherein the compensating the gas pressure sequences of the time windows based on the whole temperature sequences of the time windows to obtain a standard gas pressure sequence at a standard temperature, and extracting a gas pressure trend and a confidence of the gas pressure trend from the standard gas pressure sequence comprises: aligning the whole temperature sequence and the air pressure sequence based on time points to obtain multiple groups of basic data , wherein, Indicating a point in time The corresponding air pressure is used for controlling the air pressure, Indicating a point in time A corresponding temperature; For each point in time Based on temperature To air pressure Compensating to obtain equivalent air pressure of 20 DEG C And an equivalent air pressure sequence, wherein the mathematical expression of the equivalent air pressure is: sliding window least square method based slope of equivalent air pressure sequence of current time window Determining coefficients The slope is adjusted to As a trend of air pressure and to determine the coefficient As confidence.
  3. 3. The method for monitoring the gas protection medium voltage switchgear according to claim 1, wherein the method for constructing the contact point temperature rise prediction model comprises the following steps: acquiring a plurality of temperature rise data samples of the switch cabinet in a healthy state, wherein the temperature rise data samples comprise busbar load and temperature rise value samples; Constructing a basic mathematical expression of a contact point temperature rise prediction model, wherein the basic mathematical expression is as follows: in the formula, Represent the first The temperature rise coefficient of each contact point, The load of the bus bar is indicated, Represent the first Temperature rise at the individual contact points, Representing the offset; substituting a plurality of temperature rise data samples into the basic mathematical expression and combining least square fitting to obtain a temperature rise coefficient And a contact point temperature rise prediction model.
  4. 4. The method for monitoring a gas protection medium voltage switchgear according to claim 1, wherein calculating theoretical temperature rise of a contact point based on a pre-constructed contact point temperature rise prediction model and bus loads of a plurality of time windows, calculating residuals of the theoretical temperature rise and actual temperature rise of the contact point of the plurality of time windows, and extracting a temperature rise abnormality index of the contact point corresponding to a maximum residual of each time window comprises: For each time window, the bus bar load at each time point is calculated Substituting the theoretical temperature rise of the contact points into the contact point temperature rise prediction model of the contact points to obtain the theoretical temperature rise of the contact points at each time point ; For each contact point, an average value of theoretical temperature rise in each time window is calculated And the average value of the actual temperature rise ; For the same time window, calculate the average value of theoretical temperature rise And the average value of the actual temperature rise Residual error Wherein: in the formula, Representing a time window index; At each point in time, the maximum residual is extracted from the references of the multiple points of contact And based on the maximum residual Calculating the temperature rise abnormality index of each time window Wherein, the mathematical expression of the temperature rise abnormality index is: in the formula, Representing maximum residual error Standard deviation of the residual error of the corresponding contact point in a healthy state.
  5. 5. The method for monitoring a gas-shielded medium voltage switchgear according to claim 1, wherein the mathematical expression of the correlation is: in the formula, A sequence of air pressure trends is indicated, Representing the mean value of the air pressure trend sequence, The temperature rise abnormality index sequence is represented, Represents the average value of the temperature rise abnormality index sequence, Is the sequence length.
  6. 6. The method for monitoring a gas-shielded medium-voltage switchgear according to claim 1, wherein the method for constructing the fusion model comprises the following steps: Acquiring historical operation data samples of a plurality of historical time periods of the gas protection medium-voltage switch cabinet, wherein the historical operation data samples comprise an integral temperature sequence sample, a gas pressure sequence sample, a busbar load sequence sample and an actual temperature rise sequence sample of a plurality of contact points; Extracting a barometric trend sample, a confidence coefficient sample of barometric trend, a temperature rise abnormality index sample, an actual temperature rise average value sample and a correlation sample from the historical operation data sample, and constructing a feature vector sample based on the barometric trend sample, the confidence coefficient sample of barometric trend, the temperature rise abnormality index sample, the actual temperature rise average value sample and the correlation sample; Constructing a Bayesian diagnosis model and a DS evidence model based on feature vector samples of various types of labels, and fusing the output of the Bayesian diagnosis model and the output of the DS evidence model, wherein the mathematical expression of the fused output is as follows: in the formula, Indicates the output type tag determination result, As a first weight to be used, As a result of the second weight being set, As a result of the third weight being given, Represent the first The number of types of labels to be used, Features representing the output of the Bayesian diagnostic model Is a type label Is used to determine the posterior probability of (1), Evidence-for-type tags representing DS evidence model output Is used to determine the minimum confidence value of the (c), Evidence pair type label for representing DS evidence model output Is the highest confidence value of (1).
  7. 7. The method for monitoring a gas-shielded medium voltage switchgear according to claim 6, wherein constructing a bayesian diagnostic model based on feature vector samples of a plurality of types of labels comprises: Calculating a priori probabilities for each type of tag based on tag categories of a plurality of historical operational data samples Wherein, the mathematical expression of the prior probability is: in the formula, Indicating the number of total samples to be taken, Represent the first The number of samples of the individual type tags, Represent the first Probability of the individual type tag; building likelihood functions for feature vector samples corresponding to each type of tag ; Based on the prior probability And the likelihood function A Bayesian diagnosis model is constructed, wherein the mathematical expression of the Bayesian diagnosis model is as follows: in the formula, The posterior probability is represented by the probability of a posterior, Represent the first Likelihood functions for each type of tag, Represent the first The prior probability of the individual type tag.
  8. 8. The method for monitoring a gas-shielded medium voltage switchgear according to claim 6, wherein constructing a DS evidence model based on feature vector samples of a plurality of types of tags comprises: Constructing a pressure evidence BPA function, a temperature evidence BPA function and a related evidence BPA function, wherein the mathematical expressions of the pressure evidence BPA function, the temperature evidence BPA function and the related evidence BPA function are as follows: in the formula, Indicating the degree of trust in the gas leakage, Indicating the degree of trust in overheat of the contacts, The degree of trust of the composite fault is indicated, And Are all the weight coefficients of the two-dimensional space model, Indicating the trend of the air pressure for the target time period, For the barometric pressure trend threshold generated based on the feature vector samples, Indicating the abnormality index of the temperature rise for the target period, Indicating that a temperature rise abnormality index threshold is generated based on the eigenvector samples, The correlation coefficient representing the target period of time, Is an exponential function; confidence of the gas leakage And the contact overheat trust degree Fusion is carried out to obtain the gas-overheat trust degree The gas-overheat trust degree is adjusted Confidence with the composite fault Fusing to obtain fused trust Wherein the fusion process considers the collision coefficients; based on the fused trust level Computing trust functions Likelihood function Wherein the trust function And the likelihood function The mathematical expression of (2) is: in the formula, Representation type tag A corresponding set of fault states, Representation type tag A subset of the corresponding set of fault conditions, Representing the confidence level of the subset.
  9. 9. The method for monitoring a gas-shielded medium voltage switchgear of claim 1, further comprising: and when the monitoring result comprises any fault type, sending the monitoring result to a target object.
  10. 10. A monitoring system for a gas-shielded medium voltage switchgear, comprising: the system comprises an acquisition module, a control module and a control module, wherein the acquisition module is used for acquiring an overall temperature sequence, an air pressure sequence, a busbar load sequence and an actual temperature rise sequence of a plurality of contact points, which are acquired in a target time period in a gas protection medium-voltage switch cabinet, wherein the target time period comprises a plurality of time points of target duration before a current time point; The system comprises a target time period, an air pressure analysis module, a comparison module and a comparison module, wherein the target time period is divided into a plurality of time windows; The temperature rise analysis module is used for calculating theoretical temperature rise of the contact point based on a pre-constructed contact point temperature rise prediction model and bus loads of a plurality of time windows, calculating residual errors of the theoretical temperature rise and actual temperature rise of the contact point of the time windows, extracting a temperature rise abnormality index of the contact point corresponding to the maximum residual error of each time window, and calculating actual temperature rise average values of the contact points corresponding to the current time window; the system comprises a correlation extraction module, a temperature rise abnormality index sequence, a correlation calculation module and a correlation calculation module, wherein the correlation extraction module is used for constructing an air pressure trend sequence based on air pressure trends of a plurality of time windows, constructing a temperature rise abnormality index sequence based on temperature rise abnormality indexes of the time windows, and calculating the correlation of the air pressure trend sequence and the temperature rise abnormality index sequence; The fusion monitoring module is used for constructing a feature vector based on the air pressure trend of the current time window, the confidence coefficient of the air pressure trend of the current time window, the temperature rise abnormality index of the current time window, the actual temperature rise average value of the current time window and the correlation degree, and inputting the feature vector into a fusion model constructed in advance to obtain a monitoring result, wherein the fusion model fuses the Bayesian diagnosis model and the DS evidence theory.

Description

Monitoring method and system for gas-shielded medium-voltage switch cabinet Technical Field The application relates to the technical field of switch cabinet monitoring, in particular to a monitoring method and a system for a gas-shielded medium-voltage switch cabinet. Background The gas-insulated metal-enclosed switchgear, commonly called a gas-filled cabinet (C-GIS), has been widely used in the field of medium voltage distribution such as urban power network, rail transit, industrial factory buildings and data centers due to its compact structure, strong environmental adaptability and high reliability. The equipment realizes full insulation and full sealing by sealing high-voltage conductive parts such as a breaker, a disconnecting switch, a bus and the like in a stainless steel or aluminum alloy gas chamber filled with insulating gas with lower pressure (such as dry air, nitrogen, SF 6 or mixed gas thereof), and greatly improves safety and maintenance-free performance. In the prior art, for monitoring of gas conditions, mechanical density relays or digital density transmitters are commonly used. And alarming is realized by a mode of fixing a threshold value. Firstly, the alarm is insensitive to slow and tiny leakage trend based on a fixed threshold value, the alarm is triggered after the leakage quantity is accumulated to a certain degree, the pre-warning is lost, secondly, the gas parameter is generally independently monitored in the prior art, the non-uniformity influence on the whole temperature field of the air chamber caused by local overheating in the equipment is not considered, the pressure compensation based on single-point temperature measurement is caused by uneven temperature distribution, and erroneous judgment is possibly generated. In addition, for monitoring primary loop connection points (such as bus connection points and breaker contacts), the traditional means mainly rely on manual regular inspection through an observation window by using a thermal infrared imager or monitoring the absolute temperature of key points by using a wireless temperature measuring sensor which is installed in a discrete manner. The former can not realize real-time on-line monitoring, is limited by a patrol period, is difficult to capture sudden or rapidly-developed heat defects, and the latter can realize on-line monitoring, but is only provided with a simple upper temperature limit alarm. Such methods fail to effectively strip the significant impact of load current fluctuations on contact temperature rise, potentially triggering false alarms due to load increases, or masking the potential for substantial increases in contact resistance due to lower loads. More importantly, the existing contact point temperature monitoring system and the gas state monitoring system are independent, and the data and the alarm information are not related. In actual operation, the aging of the air chamber seal may cause micro leakage, and may change local heat dissipation conditions or introduce moisture due to the reduced sealing performance, exacerbating oxidation and corrosion of adjacent electrical connection points, and forming a correlation failure of "air leakage" and "overheating". The existing isolated monitoring mode can not identify the composite fault mode at all, and is not beneficial to root cause analysis and accurate maintenance strategy formulation. Disclosure of Invention Therefore, the application aims to provide a monitoring method and a system for a gas-shielded medium-voltage switch cabinet, which are used for solving the problems in the background technology. In order to achieve the above purpose, the present application adopts the following technical scheme: the application discloses a monitoring method of a gas-shielded medium-voltage switch cabinet, which comprises the following steps: Acquiring an overall temperature sequence, an air pressure sequence, a busbar load sequence and an actual temperature rise sequence of a plurality of contact points, which are acquired in a target time period in the gas protection medium-voltage switch cabinet, wherein the target time period comprises a plurality of time points of target duration before a current time point; the target time period is divided into a plurality of time windows, the air pressure sequences of the time windows are compensated based on the whole temperature sequences of the time windows to obtain a standard air pressure sequence at the standard temperature, and the air pressure trend and the confidence coefficient of the air pressure trend are extracted from the standard air pressure sequence; calculating theoretical temperature rise of the contact point based on a pre-constructed contact point temperature rise prediction model and bus loads of a plurality of time windows, calculating residual errors of the theoretical temperature rise and the actual temperature rise of the contact point of the time windows, extracting a temperature rise abnormal