Search

CN-121254012-B - Cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and implementation method

CN121254012BCN 121254012 BCN121254012 BCN 121254012BCN-121254012-B

Abstract

The invention relates to the technical field of cable insulation monitoring and fault diagnosis, and particularly discloses a cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and an implementation method. The system comprises a data acquisition module, an interference analysis module, an attenuation analysis module, a cluster intervention module and a reversible conversion time, wherein the data acquisition module acquires insulation resistance and load current parameters and performs null registration to output a load resistance data set, the interference analysis module screens insulation analysis segments and judges latent damage accumulation through energy accumulation analysis, the attenuation analysis module calculates an attenuation reversible value through impact strength and recovery efficiency coupling analysis and judges whether an attenuation early warning signal is triggered or not, the cluster intervention module divides a global impact attenuation segment into space clusters after triggering early warning, and a reversible conversion prediction model is built by combining the reversible parameters and the impact parameters to output the attenuation conversion time. The method is used for realizing the cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis. The invention improves the pre-diagnosis accuracy through multi-parameter coupling and global cluster analysis, and provides a basis for cable operation and maintenance.

Inventors

  • RUAN FA
  • RUAN XINYU
  • QU MINGFEI

Assignees

  • 安徽帕维尔智能技术有限公司

Dates

Publication Date
20260505
Application Date
20251013

Claims (10)

  1. 1. The cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system is characterized by comprising the following modules: The data acquisition module is used for acquiring insulation parameters of the cable monitoring section and carrying out space-time registration analysis to output a load-resisting data set; The interference analysis module classifies the monitoring segments based on the load-blocking data set to obtain sudden-rise and sudden-fall type monitoring segments and extract micro-fluctuation characteristics, screens the insulation analysis segments based on the micro-wave dynamic characteristics, performs latent damage analysis on the insulation analysis segments, and judges whether the insulation analysis segments have latent damage accumulation or not; the attenuation analysis module is used for carrying out attenuation identification on the insulation analysis section based on the load current if the latent damage is accumulated, obtaining an impact attenuation section, carrying out attenuation reversibility analysis on the impact attenuation section, obtaining an attenuation reversibility value and judging whether to trigger an attenuation early warning signal; And the cluster intervention module is used for dividing the global imperial punching weakening section into a plurality of space clusters if the attenuation early warning signal is triggered, constructing a reversible prediction model, acquiring reversible parameters and impact parameters of each space cluster, inputting the model, outputting attenuation conversion time and performing fault intervention on the space clusters.
  2. 2. The system for dynamically monitoring and pre-diagnosing faults on cable insulation according to claim 1, wherein the space-time registration analysis is performed in the following manner: Acquiring a load current value and an insulation resistance value of each monitoring segment as insulation parameters, and simultaneously extracting space coordinate codes and sampling time stamps of the load current value and the insulation resistance value; according to the same space coordinate code, matching the insulation resistance value and the load current value of the monitoring segmented cable according to the insulation resistance value and the load current value of the same sampling time stamp to form a quaternary data set of time-space-insulation resistance value-load current value; And acquiring quaternary data sets of all monitoring segments at N sampling time stamps, and constructing a load-resisting data set.
  3. 3. The system for dynamically monitoring and diagnosing faults on cable insulation multiple parameters according to claim 1 is characterized in that the mode of extracting the microwave dynamic characteristics is as follows: acquiring a steady-state interval preset by the load current, and determining a recovery period of a recovery stage based on the steady-state interval of the load current; extracting microwave dynamic frequency and direction fluctuation ratio in a recovery period, and taking the microwave dynamic frequency and the direction fluctuation ratio as microwave dynamic characteristics; The direction fluctuation ratio comprises a positive fluctuation ratio and a negative fluctuation ratio.
  4. 4. The system for dynamically monitoring and diagnosing faults on cable insulation multiple parameters according to claim 3, wherein the mode of extracting the direction fluctuation ratio is as follows: Acquiring an insulation resistance reference value, constructing a fluctuation start-stop judging condition and an effective fluctuation condition based on the insulation resistance reference value, and judging effective fluctuation when the insulation resistance value of the monitoring section meets the fluctuation start-stop judging condition and the effective fluctuation condition; if the insulation resistance value in the effective fluctuation in the recovery period is lower than the insulation resistance reference value, marking the effective fluctuation as negative deviation fluctuation, and if the effective fluctuation is higher than the insulation resistance reference value, marking the effective fluctuation as positive deviation fluctuation; calculating the proportion of the negative deviation fluctuation times of the insulation resistance value in the recovery period to all fluctuation times as a negative fluctuation ratio; and calculating the proportion of the forward deviation fluctuation times of the insulation resistance value in the recovery period to the total fluctuation times as a forward fluctuation ratio.
  5. 5. The system for dynamically monitoring and pre-diagnosing faults of cable insulation according to claim 1, wherein the method for analyzing the latent damage is as follows: If the negative fluctuation ratio of the monitoring segments for M continuous monitoring periods is higher than the positive fluctuation ratio, marking the monitoring segments as insulation analysis segments; For each negative effective fluctuation of the insulation analysis section in the recovery period, obtaining the negative fluctuation energy of the recovery period through a negative fluctuation energy formula; Calculating the total negative fluctuation energy of the recovery period to obtain period accumulated energy; And (3) obtaining the periodic energy of the M monitoring period insulation analysis sections, performing linear regression fitting, and if the slope of the fitting is positive, judging that the periodic accumulated energy of the M monitoring period insulation analysis sections has an increasing trend, and judging that the latent damage accumulation exists.
  6. 6. The system for dynamically monitoring and pre-diagnosing faults of cable insulation according to claim 1, wherein the weakening identification is performed in the following manner: Calculating an absolute difference value between an insulation resistance reference value and an extreme value of the load current as an absolute variation; acquiring absolute variation of the insulation analysis section in M monitoring periods, and constructing an absolute variation sequence; performing cluster analysis on the absolute change sequence, and dividing the absolute change amount into different intensity levels based on a cluster analysis result; calculating the fluctuation frequency mean value of each intensity level in the current monitoring period, and carrying out KL divergence calculation with a normal fluctuation reference value to obtain the deviation degree of each intensity level; And constructing a weakening judgment criterion based on the deviation degree, and if the deviation degree of each intensity level of the insulation analysis section meets the weakening judgment criterion, judging that the insulation analysis section is a blunting weakening section.
  7. 7. The system for dynamically monitoring and pre-diagnosing faults of cable insulation according to claim 1, wherein the attenuation reversibility analysis is performed in the following manner: Acquiring a recovery efficiency-variation data set and a historical recovery efficiency-variation data set; Fitting the imperial punching weakened section in a recovery efficiency-variable quantity data set and a historical recovery efficiency-variable quantity data set by adopting a nonlinear regression algorithm to obtain a current curve and a historical curve; Performing feature extraction processing on the two curves to obtain a recovery attenuation coefficient and a slope degradation ratio; summing and averaging the recovered attenuation coefficients of all the intensity levels to obtain an attenuation coefficient average value; And constructing a proportional equation, and inputting the attenuation coefficient mean value and the slope degradation ratio into the proportional equation to obtain an attenuation reversible value.
  8. 8. The system for dynamically monitoring and pre-diagnosing faults of cable insulation according to claim 7, wherein the feature extraction processing is performed in the following manner: extracting characteristic impact strength points of the two curves at each strength level; calculating the difference ratio of the recovery efficiency and the historical recovery efficiency of the current curve and the historical curve before the characteristic impact strength point of each strength grade, and obtaining the recovery attenuation coefficient of each strength grade; Calculating the average slope of the current curve from the middle intensity level to the high intensity level as the current intensity slope; Calculating the average slope of the historical curve from the medium intensity level to the high intensity level as the historical intensity slope; The absolute duty ratio of the current intensity slope to the historical intensity slope is calculated as the slope degradation ratio.
  9. 9. The system for dynamically monitoring and diagnosing faults with multiple parameters on cable insulation according to claim 1, wherein the mode of outputting the attenuation conversion time is as follows: For each space cluster, calculating the total number of the imperial punching weakened sections in each cluster, namely the total number of the clusters; Acquiring the number of irreversible attenuation of the impact attenuation sections in each space cluster, and carrying out ratio processing on the number of irreversible attenuation sections and the total number of the clusters to obtain a cluster non-inverse ratio; Taking a cluster non-inverse ratio and an attenuation reversible value of the space cluster as reversible parameters; Taking the absolute variation of load impact and the negative fluctuation ratio of each impact resisting weakening section in the space cluster as impact parameters; And constructing a reversible transformation prediction model based on the reversible parameters and the impact parameters, and outputting the attenuation transformation time for transforming all the reversible attenuation imperfection sections in the space cluster into the irreversible imperfection sections.
  10. 10. A method for realizing the cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis is characterized by being used for realizing the cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system according to any one of claims 1-9.

Description

Cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and implementation method Technical Field The invention relates to the technical field of cable insulation monitoring and fault diagnosis, in particular to a cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and an implementation method. Background The cable is used as a core transmission carrier of the power system, the insulation performance of the cable determines the stability and safety of power transmission, and along with the capacity improvement and the operation period extension of the power system, the cable insulation is easily affected by load impact and environmental factors to cause latent damage, and if the cable cannot be timely identified and managed, power faults are easily generated to influence power supply. The prior art lacks global clustering analysis on scattered high-risk monitoring segments, is difficult to allocate resources in a targeted manner in the operation and maintenance process, is easy to generate a management and control blind area or excessive operation and maintenance, and has the overall early warning and management and control efficiency to be improved. Therefore, the invention provides a cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and an implementation method. Disclosure of Invention The invention aims to provide a cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system and an implementation method thereof so as to solve the background problem. The aim of the invention can be achieved by the following technical scheme: the cable insulation multi-parameter dynamic monitoring and fault pre-diagnosis system comprises the following modules: The data acquisition module is used for acquiring insulation parameters of the cable monitoring section and carrying out space-time registration analysis to output a load-resisting data set; The interference analysis module classifies the monitoring segments based on the load-blocking data set to obtain sudden-rise and sudden-fall type monitoring segments and extract micro-fluctuation characteristics, screens the insulation analysis segments based on the micro-wave dynamic characteristics, performs latent damage analysis on the insulation analysis segments, and judges whether the insulation analysis segments have latent damage accumulation or not; the attenuation analysis module is used for carrying out attenuation identification on the insulation analysis section based on the load current if the latent damage is accumulated, obtaining an impact attenuation section, carrying out attenuation reversibility analysis on the impact attenuation section, obtaining an attenuation reversibility value and judging whether an attenuation signal is triggered or not; And the cluster intervention module is used for dividing the global imperial punching weakening section into a plurality of space clusters if the attenuation early warning signal is triggered, constructing a reversible prediction model, acquiring reversible parameters and impact parameters of each space cluster, inputting the model, outputting attenuation conversion time and performing fault intervention on the space clusters. As a further aspect of the invention, the spatiotemporal registration analysis is performed by: Acquiring a load current value and an insulation resistance value of each monitoring segment as insulation parameters, and simultaneously extracting space coordinate codes and sampling time stamps of the load current value and the insulation resistance value; according to the same space coordinate code, matching the insulation resistance value and the load current value of the monitoring segmented cable according to the insulation resistance value and the load current value of the same sampling time stamp to form a quaternary data set of time-space-insulation resistance value-load current value; And acquiring quaternary data sets of all monitoring segments at N sampling time stamps, and constructing a load-resisting data set. The method for extracting the microwave dynamic characteristics comprises the following steps of: acquiring a steady-state interval preset by the load current, and determining a recovery period of a recovery stage based on the steady-state interval of the load current; extracting microwave dynamic frequency and direction fluctuation ratio in a recovery period, and taking the microwave dynamic frequency and the direction fluctuation ratio as microwave dynamic characteristics; The direction fluctuation ratio comprises a positive fluctuation ratio and a negative fluctuation ratio. The method for extracting the direction fluctuation ratio comprises the following steps of: Acquiring an insulation resistance reference value, constructing a fluctuation start-stop judging condition and an effective fluctuation condition based on the insulation resistance reference value, and