Cable insulation abnormity and latent fault early warning method and device based on electrical quantity

Abstract

The invention discloses a cable insulation abnormity and latent fault early warning method and device based on electrical quantity, and the method comprises the steps: synchronously monitoring the zero-sequence voltage and current of a power distribution network in real time, starting a fault or disturbance analog quantity data cache when a condition is satisfied, and generating a wave recording file; feature frequency band transient fault information is extracted, ground fault judgment is completed, and the fault type is judged; recording and calculating instantaneous fault effective data, and judging whether the instantaneous grounding fault frequency in a preset time limit exceeds a threshold value or not; defining an insulation deterioration degree evaluation index, determining the evaluation grade of the fault, and giving an alarm; initially positioning a cable insulation abnormity and latent fault section; and comparing the waveform of the wave recording file, expanding the section nodes for checking, and if no error exists, carrying out latent fault early warning. The method solves the problem that the power supply reliability is reduced due to the fact that the insulation state of the power distribution network cable is difficult to monitor, many faults are disposed afterwards and unplanned power failure occurs, does not need to additionally add other equipment including a sensor, and has wide practical application value.

Description

Technical Field

[0001] The present invention belongs to the technical field of cable insulation status monitoring and fault early warning, and more specifically, relates to a cable insulation abnormality and latent fault early warning method and device based on electrical quantities. Background Art

[0002] During the normal operation of power cables, as the insulation deteriorates, a series of self-clearing transient faults will occur. If preventive maintenance and repairs are not carried out in time, the cable insulation will be broken down and a permanent grounding fault will occur. Most of these transient faults are caused by the influence of the external environment on the cable body or accessories. For example, a transient grounding fault occurs when the cable joint is corroded by water. At the same time, it is accompanied by a transient overvoltage. The energy generated by the arc at the fault point causes the water vapor inside the insulation to evaporate rapidly, developing into high-pressure steam, which extinguishes the arc at the fault grounding point in a very short time, thereby restoring the normal operation of the cable. Relevant literature shows that the duration of a typical transient fault of a power cable is very short, usually only a few milliseconds, about 1 / 4 cycle (5ms), and the first transient fault is a single-phase grounding fault. Therefore, transient faults can be regarded as a sign of deterioration of the insulation of the power cable.

[0003] As the cable insulation medium gradually deteriorates, the frequency of transient faults increases, eventually leading to cable insulation breakdown and ground faults. The first transient fault may occur weeks or months before the final breakdown of the cable insulation. In the prior art, there is a method for judging whether a line fault occurs in the line by the busbar zero-sequence voltage, and identifying whether the fault type is a transient fault or a permanent fault. However, it is unable to further accurately monitor the insulation status of the distribution network cable through multiple criteria and provide advance warning of the fault. Therefore, there is an urgent need for a cable insulation abnormality and latent fault warning method, device and system based on electrical quantity information. Summary of the invention

[0004] In order to address the deficiencies in the prior art, the present invention provides a cable insulation anomaly and latent fault warning method, device and system based on electrical quantity information, which is applicable to all small current grounding systems and has broad practical application value. It can timely monitor and discover and accurately locate transient faults, evaluate the degree of cable insulation abnormality during transient faults and issue warnings, so that power grid operation personnel can have enough time to repair or replace faulty accessories, avoid cable insulation breakdown accidents, reduce unnecessary power outages, and improve power supply reliability.

[0005] The present invention adopts the following technical solution.

[0006] A first aspect of the present invention provides a cable insulation anomaly and latent fault early warning method based on electrical quantities, comprising the following steps:

[0007] Step S1: Real-time synchronous monitoring of the zero-sequence voltage and zero-sequence current of the distribution network system. When the zero-sequence voltage or zero-sequence current meets the fault start condition, the fault or disturbance analog data cache is started and a recording file is generated at the same time. Otherwise, the zero-sequence voltage and zero-sequence current of the distribution network system are continuously monitored.

[0008] Step S2: extracting transient fault information from the characteristic frequency bands of the zero-sequence voltage and the zero-sequence current to complete ground fault identification; if there is no ground fault, uploading the wave recording file for verification and ending the step; if there is a ground fault, judging whether the fault is a permanent fault according to the transient fault information; if it is a permanent fault, directly issuing an alarm signal and ending the step; otherwise, it is classified as a transient ground fault;

[0009] Step S3: Record and calculate the valid data of instantaneous ground faults, and determine whether the number of instantaneous ground faults within a preset time limit exceeds the threshold corresponding to the preset time limit. If so, it is determined that the insulation of the cable in the downstream area of ​​this node is abnormal, an alarm signal is issued and the process jumps to step S5;

[0010] Step S4: defining the insulation degradation evaluation index and dividing the evaluation levels, and based on the valid data of the instantaneous ground fault in S3, determining the evaluation level to which the instantaneous ground fault belongs and issuing a corresponding alarm signal;

[0011] Step S5: transmitting alarm signals and recording files according to communication conditions, and preliminarily locating the cable insulation abnormality section or potential fault section;

[0012] Step S6: Compare the waveforms in the recording files of the adjacent nodes of the initially determined cable insulation abnormality or latent fault section, expand the insulation abnormality or latent fault section nodes for verification, and issue a cable insulation abnormality or latent fault warning if it is correct, otherwise, it is classified as a system disturbance.

[0013] Preferably, the fault initiation condition is the collected system zero-sequence voltage amplitude U 0 Exceeds the preset threshold U set Or zero sequence voltage amplitude change ΔU 0 Exceeds the preset threshold ΔU set Or zero sequence current amplitude I 0max Exceeds the preset threshold I set Or node zero-sequence current amplitude change ΔI 0max Exceeds the preset threshold ΔI 0set .

[0014] Preferably, in step S1, Beidou or GPS or IRIG-B code or SNTP is used to synchronize the time of the synchronous monitoring process.

[0015] Preferably, in step S1, the recording file types include transient recording, continuous recording and dynamic recording; the recording file adopts the standard COMTRADE format, the transient recording sampling rate f is not less than 12.8kHz, and at least contains 80ms before and 100ms data after the trigger moment; the continuous recording forms a file per minute, and the sampling rate is not less than 1.6kHz; the dynamic recording rate is not less than 25 times / second.

[0016] Preferably, in step S2, Hilbert transform is used to perform time-frequency domain processing on the zero-sequence voltage and zero-sequence current to extract transient fault information in a characteristic frequency band.

[0017] Preferably, in step S2, the lower limit of the characteristic frequency band range is 150 Hz and the upper limit is 2000 Hz.

[0018] Preferably, in step S2, whether a ground fault exists is determined locally by any one of a DC offset method, a transient power direction method, a first half-wave method, and a three-phase current imbalance method.

[0019] Preferably, in step S3, the preset time limit includes a short time limit or a long time limit, the short time limit is set to 1 minute, and the short time limit instantaneous ground fault number threshold is set to 3 times; the long time limit is set to 3 minutes, and the long time limit instantaneous ground fault number threshold is set to 5 times.

[0020] Preferably, the valid data of the instantaneous ground fault includes the fault duration T D , zero mode voltage effective value U 0rms , Fault interval time T DS , Number of instantaneous ground faults within a short time limit N ST , Long-term instantaneous ground fault fault N LT 、Number of ground faults within 24 hours tD , Fault impact elimination time T CL , Permanent ground fault sign F PG , the number of faults within the elimination time N fY , Monthly failure rate change rate ΔV fm ;

[0021] In step S3, recording and calculating the effective data of the instantaneous ground fault specifically includes:

[0022] Fault duration T D : Taking the sampling value u of the zero-sequence voltage of the distribution network system 0 Exceeding the threshold value U for the first time set The starting time is less than U set And it does not exceed U again in the next power frequency cycle set The moment of is the termination moment;

[0023] Zero mode voltage effective value U 0rms : Record the zero-mode voltage effective value U 0rms , expressed as the following formula,

[0024]

[0025] Where:

[0026] u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system;

[0027] N is the fault duration T D The number of sampling points within

[0028] Fault interval time T DS : The time interval between this ground fault and the previous ground fault;

[0029] Number of instantaneous ground faults within a short time limit N ST : The number of instantaneous ground faults detected within a short time limit;

[0030] Long-term instantaneous ground fault fault N LT : The number of instantaneous ground faults detected within a long time limit;

[0031] Number of ground faults within 24 hours N tD : The number of insulation failures within 24 hours;

[0032] Fault impact elimination time T CL :Calculate the time T for eliminating the impact of the fault CL , expressed as the following formula,

[0033] T CL =T max -T ki

[0034] Where:

[0035] T max The maximum impact time of the fault is set according to the maintenance cycle;

[0036] T ki is the time interval between the current time and the i-th ground fault;

[0037] Permanent ground fault flag F PG : Ground fault duration T D If it is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1;

[0038] Eliminate the number of faults within the time N fY : Fault influence elimination time T CL Number of instantaneous faults occurring within

[0039] Monthly failure rate change rate ΔV fm : Change rate of the number of failures in a natural month compared with the previous month.

[0040] Preferably, step S4 specifically includes:

[0041] Step S401: The insulation degradation evaluation index D is set according to the set threshold. i The size range of is divided into N continuous evaluation levels;

[0042] Step S402, selecting M valid data from the valid data of the instantaneous ground fault as evaluation factors, and performing fuzzy processing on each evaluation factor using a membership function;

[0043] Step S403, calculating the membership degree r of each evaluation factor to each evaluation level after fuzzy processing ij , with membership degree r ij As matrix elements, establish the fuzzy relationship matrix R;

[0044] Step S404, using the weight matrix W and the fuzzy relationship matrix R to calculate the node insulation degradation degree evaluation index D i For the membership matrix B of each evaluation level, the insulation degradation degree evaluation index D is obtained according to the maximum membership principle. i If there are multiple maximum elements with the same value in the membership matrix B, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements;

[0045] Step S405, set D i The inference calculation supplementary rules for D i Adjust and determine the final insulation degradation evaluation index D i value;

[0046] Step S406: Each node is evaluated based on the final insulation degradation degree evaluation index D i , determine the evaluation level of the instantaneous ground fault and output the corresponding alarm signal.

[0047] Preferably, D is set i The size range is [0, 1], and each threshold is set, including 0.5, 0.75, and 0.85. According to each threshold, the insulation degradation evaluation index D i It is divided into four evaluation levels, including emergency alarm, severe alarm, general alarm, and latent fault; the size range of emergency alarm is [0.85, 1], the size range of severe alarm is [0.75, 0.85), the size range of general alarm is [0.5, 0.75), and the size range of latent fault is [0, 0.5).

[0048] Preferably, the fault duration T is selected D , zero mode voltage effective value U 0rms And the fault interval time T DS As an evaluation factor, the instantaneous ground fault duration T D The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0049]

[0050] Where:

[0051] T D ' represents T after gelatinization D ;

[0052] The effective value of zero-mode voltage U 0rms The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0053]

[0054] Where:

[0055] U 0rms ′ U after gelatinization 0rms ;

[0056] The time interval between consecutive instantaneous faults is T DS The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0057]

[0058] Where:

[0059] T DS ' represents T after gelatinization DS .

[0060] Preferably, in step S403, the degree of membership is calculated based on the descending semi-trapezoidal function, which is expressed by the following formula:

[0061]

[0062] Where:

[0063] r ij represents the membership of the i-th evaluation factor to the j-th evaluation level;

[0064] S j , S j+1 Represents two adjacent thresholds of the jth evaluation level;

[0065] Ci Represents the calculated value of the membership function of the i-th evaluation factor;

[0066] Calculate r ij , we get the fuzzy relationship matrix R, which is expressed as follows:

[0067] R=[r ij ] M*N

[0068] Where:

[0069] M is the number of evaluation factors;

[0070] N is the number of evaluation levels.

[0071] Preferably, in step S404, the fault duration T is allocated D , zero mode voltage effective value U 0rms And the fault interval time T DS The weight of the fault duration T D , zero mode voltage effective value U 0rms And the fault interval time T DS The weights of are used to construct the weight matrix W, then B = W × R.

[0072] Preferably, in step S405, three inference calculation supplementary rules are set, and the three inference calculation supplementary rules are:

[0073] 1) If the number of instantaneous ground faults within 24 hours is N tD If it is greater than 24 times, output 1;

[0074] 2) If the continuous instantaneous fault interval T in the historical fault data of the last month is DS If the difference is decreasing, output 1;

[0075] 3) If the most recent fault affects the elimination time T CL If it is greater than 30 days, +0.2 is added to the evaluation value result of step S404.

[0076] Preferably, step S6 specifically includes:

[0077] Step S601, select reference fault recording trigger time t 0 The triggering time of other fault recordings is t i ,Using the wide area clock synchronization, the sampling points of other recorded data are shifted, the fault recorded data in the same area are aligned, and the similarity of zero-sequence current waveforms of adjacent nodes after alignment is compared;

[0078] Step S602, perform zero-sequence current waveform similarity check on the transient recording of the extended node. If the transient recording waveform of the extended node fails to cover the waveform at the initial moment of the fault, then use the continuous recording file of the minute at which the current transient recording file is triggered to check the zero-sequence current waveform similarity. If the fault is subsequently inverted, then combine the dynamic recording file to check the zero-sequence current waveform similarity.

[0079] Step S603, judging the latent fault or system disturbance according to the similarity of the zero-sequence current waveforms of the adjacent nodes and the extended nodes.

[0080] The second aspect of the present invention provides a cable insulation anomaly and latent fault early warning device based on electrical quantities, which has the three remote control and protection functions of the distribution automation station terminal, including: a real-time core, a management core, a communication module, a measurement and control module, an input and output module and a protection module;

[0081] The real-time core includes a ground fault determination unit, a wave recording and index data calculation unit;

[0082] The ground fault determination unit is used to extract transient fault information from the characteristic frequency bands of zero-sequence voltage and zero-sequence current to complete ground fault determination;

[0083] The wave recording and index data calculation unit is used to start the fault or disturbance analog data cache, generate a wave recording file, record and calculate the effective data of the instantaneous ground fault, and determine whether the number of instantaneous ground faults within a preset time limit exceeds a threshold value corresponding to the preset time limit;

[0084] The measurement and control module is used to synchronously monitor the zero-sequence voltage and zero-sequence current of the power distribution network system in real time;

[0085] The binary input and output module is used to collect and output binary quantities;

[0086] The protection module is used to determine whether the zero-sequence voltage or zero-sequence current meets the fault start condition and issue an alarm signal;

[0087] The management core includes a historical fault data storage unit, an insulation degradation index calculation and statistical data analysis unit, and a device status monitoring unit;

[0088] The historical fault data storage module is used to store the recording files and calculation results of the recording and index data calculation unit;

[0089] The insulation degradation index calculation and statistical data analysis unit is used to define the insulation degradation degree evaluation index and divide the evaluation level, and determine the evaluation level to which the instantaneous ground fault belongs based on the valid data of the instantaneous ground fault in the historical fault data storage module;

[0090] The device status monitoring unit is used to monitor the operation of the main body, the channel link status and the data abnormality alarm;

[0091] The communication module is used to transmit alarm information and fault recording according to communication conditions;

[0092] The real-time core and the management core operate as dual cores, and data interaction is achieved through the data bus;

[0093] The device also includes: a power-off management core, which is used to transfer and store data when the device is powered off; and a power-on real-time core, which is used to read the most recently transferred and stored data.

[0094] Preferably, the fault start condition in the protection module is the amplitude U of the collected system zero-sequence voltage 0 Exceeds the preset threshold U set , or the amplitude change of zero-sequence voltage ΔU 0 Exceeds the preset threshold ΔU set , or the amplitude of the system zero-sequence current I 0max Exceeds the preset threshold I set , or the amplitude change of the node zero-sequence current ΔI 0max Exceeds the preset threshold ΔI 0set .

[0095] Preferably, the preset time limit in the wave recording and index data calculation unit includes a short time limit or a long time limit, the short time limit is set to 1 minute, and the threshold corresponding to the short time limit is 3 times; the long time limit is set to 3 minutes, and the threshold corresponding to the long time limit is 5 times;

[0096] The valid data of instantaneous ground fault in the waveform recording and index data calculation unit include the fault duration T D , zero mode voltage effective value U 0rms , Fault interval time T DS , Number of instantaneous ground faults within a short time limit N ST , Long-term instantaneous ground fault fault N LT 、Number of ground faults within 24 hours Nt D , Fault impact elimination time T CL , Permanent ground fault sign F PG , the number of faults within the elimination time N fY , Monthly failure rate change rate ΔV fm ;

[0097] The recording and index data calculation unit records and calculates the effective data of the instantaneous ground fault, specifically including:

[0098] Fault duration T D : Taking the sampling value u of the zero-sequence voltage of the distribution network system 0 Exceeding the threshold value U for the first time setThe starting time is less than U set And it does not exceed U again in the next power frequency cycle set The moment of is the termination moment;

[0099] Zero mode voltage effective value U 0rms : Record the zero-mode voltage effective value U 0rms , expressed as the following formula,

[0100]

[0101] Where:

[0102] u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system;

[0103] N is the fault duration T D The number of sampling points within

[0104] Fault interval time T DS : The time interval between this ground fault and the previous ground fault;

[0105] Number of instantaneous ground faults within a short time limit N ST : The number of instantaneous ground faults detected within a short time limit;

[0106] Long-term instantaneous ground fault fault N LT : The number of instantaneous ground faults detected within a long time limit;

[0107] Number of ground faults within 24 hours N tD : The number of insulation failures within 24 hours;

[0108] Fault impact elimination time T CL : Fault influence elimination time T CL The calculation formula is,

[0109] T CL =T max -T ki

[0110] Where:

[0111] T max The maximum impact time of the fault is set according to the maintenance cycle;

[0112] T ki is the time interval between the current time and the i-th ground fault;

[0113] Permanent ground fault flag F PG : Ground fault duration T DIf it is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1;

[0114] Eliminate the number of faults within the time N fY : Fault influence elimination time T CL Number of instantaneous faults occurring within

[0115] Monthly failure rate change rate ΔV fm : Change rate of the number of failures in a natural month compared with the previous month.

[0116] Preferably, in the insulation degradation index calculation and statistical data analysis unit, determining the evaluation level to which the instantaneous ground fault belongs specifically includes:

[0117] According to the set threshold, the insulation degradation evaluation index D i The size range is divided into N continuous evaluation levels; M valid data are selected from the valid data of instantaneous grounding fault as evaluation factors, and fuzzy processing is performed using membership functions respectively; the membership r of each evaluation factor to each evaluation level after fuzzy processing is calculated ij , with membership degree r ij As matrix elements, the fuzzy relationship matrix R is established; the node insulation degradation degree evaluation index D is calculated using the weight matrix W and the fuzzy relationship matrix R i For the membership matrix B of each evaluation level, the insulation degradation degree evaluation index D is obtained according to the maximum membership principle. i If there are multiple maximum elements with the same value in the membership matrix B, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements; set D i The inference calculation supplementary rules for D i Adjust and determine the final insulation degradation evaluation index D i Value; Each node is evaluated according to the final insulation degradation degree index D i , determine the evaluation level of the instantaneous ground fault.

[0118] Preferably, the communication module shares alarm information with adjacent devices via a horizontal communication channel, or transmits alarm information and fault recording to a distribution automation master station via a vertical communication channel.

[0119] The third aspect of the present invention provides a cable insulation anomaly and latent fault warning system based on electrical quantities, comprising: a distribution automation master station and multiple devices as described in the second aspect of the present invention; the devices and the devices and the distribution automation master station are interconnected based on horizontal communication channels or vertical communication channels to form a system, which performs segmented positioning and early warning of insulation anomalies and latent faults of distribution network cables.

[0120] A fourth aspect of the present invention provides a computer device, comprising: one or more processors;

[0121] The processor is used to store one or more programs;

[0122] When the one or more programs are executed by the one or more processors, the electrical quantity-based cable insulation anomaly and latent fault early warning method is implemented.

[0123] A fifth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed, the electrical quantity-based cable insulation anomaly and latent fault warning method is implemented.

[0124] Compared with the prior art, the beneficial effects of the present invention include at least:

[0125] (1) The present invention uses the electrical quantity information of each node in the distribution network system to realize the early warning of cable insulation abnormality and latent faults, without the need to add other equipment including sensors;

[0126] (2) The present invention evaluates the insulation status according to the frequency, duration, voltage amplitude, etc. of transient faults, and performs segment positioning based on network topology through field node information sharing or distribution automation master station. It effectively solves the problems of difficult insulation status monitoring of distribution network cables, post-fault handling of multiple faults, and unplanned power outages that reduce power supply reliability, and has broad practical application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0127] Figure 1 is a logic flow chart of a method provided according to an embodiment of the present invention;

[0128] Figure 2 is a diagram of the device module composition provided according to an embodiment of the present invention;

[0129] Figure 3 A diagram of a system having both vertical and horizontal communication is provided according to an embodiment of the present invention;

[0130] Figure 4 It is a diagram of a system having only vertical communication provided in accordance with an embodiment of the present invention. DETAILED DESCRIPTION

[0131] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0132] like Figure 1 As shown, embodiment 1 of the present invention provides a cable insulation abnormality and latent fault early warning method based on electrical quantity information, comprising the following steps:

[0133] Step S1: Real-time synchronous monitoring of system zero-sequence voltage U 0 And the current node zero sequence current I 0n When the zero-sequence voltage or zero-sequence current meets the fault start condition, the fault or disturbance analog data cache is started and a recording file is generated at the same time, otherwise the zero-sequence voltage and zero-sequence current of the distribution network system continue to be monitored.

[0134] Preferably, the fault initiation condition is the collected system zero-sequence voltage amplitude U 0 Exceeds the preset threshold U set Or zero sequence voltage amplitude change ΔU 0 Exceeds the preset threshold ΔU set Or zero sequence current amplitude I 0max Exceeds the preset threshold I set Or node zero-sequence current amplitude change ΔI 0max Exceeds the preset threshold ΔI 0set .

[0135] Preferably, the time synchronization mode of the synchronous monitoring process includes Beidou / GPS, IRIG-B code or SNTP, and the recording file types include transient recording, continuous recording and dynamic recording. 0 With the current node zero sequence current I 0n The methods include external or self-production.

[0136] Further preferably, the recording file adopts the standard COMTRADE format, the transient recording sampling rate f is not less than 12.8kHz, and at least contains data 80ms before and 100ms after the trigger moment; continuous recording forms one file per minute, the sampling rate is not less than 1.6kHz; the dynamic recording rate is not less than 25 times / second.

[0137] Step S2: extracting transient fault information from the characteristic frequency bands of the zero-sequence voltage and the zero-sequence current, and completing the ground fault identification on site; if no ground fault is identified, uploading the waveform recording file for verification and ending the step; if a ground fault is identified, continuing to determine whether it is a permanent ground fault based on the transient fault information; if so, directly issuing a permanent ground fault alarm signal and ending the step; otherwise, it is classified as a transient ground fault and continuing to execute step S3.

[0138] Preferably, Hilbert transform is used to perform time-frequency domain processing on the zero-sequence voltage and zero-sequence current to extract transient fault information in a characteristic frequency band.

[0139] Preferably, the characteristic frequency band has a lower limit of 150 Hz and an upper limit of 2000 Hz, and the impedances of both the healthy lines and the faulty lines of the system are capacitive.

[0140] Preferably, whether a ground fault exists is determined on-site by using a DC offset method, a transient power direction method, a first half-wave method, a three-phase current imbalance, and the like.

[0141] Step S3: Record and calculate the effective data of the instantaneous ground fault, including the instantaneous ground fault duration T D , zero mode voltage effective value U 0rms , continuous instantaneous fault interval time T DS , Number of instantaneous ground faults within a short time limit N ST , Number of instantaneous ground faults within a long time limit N LT 、Number of instantaneous ground faults within 24 hours tD , Fault impact elimination time T CL , Permanent ground fault sign F PG , the number of faults within the elimination time N fY , Monthly failure rate change rate ΔV fm ; Set long and short time limits, and determine the number of instantaneous ground faults N within the short or long time limit ST or N LT Whether it exceeds the threshold, if so, it is determined that the cable insulation in the downstream area of ​​this node is abnormal, an alarm signal is issued and jumps to step S5.

[0142] Preferably, the short time limit is set to 1 minute, and the short time limit instantaneous ground fault number threshold is set to 3 times; the long time limit is set to 3 minutes, and the long time limit instantaneous ground fault number threshold is set to 5 times.

[0143] Preferably, recording and calculating effective data of the instantaneous ground fault specifically includes:

[0144] Fault duration T D : Taking the sampling value u of the zero-sequence voltage of the distribution network system 0 Exceeding the threshold value U for the first time set The starting time is less than U set And it does not exceed U again in the next power frequency cycle set The moment of the fault is the termination moment; if it is an intermittent grounding fault, since most of them reignite within half a power frequency cycle after arc extinction, the entire intermittent cycle is regarded as the fault duration T D ;

[0145] Zero mode voltage effective value U 0rms :The magnitude of zero-mode voltage is inversely proportional to the grounding impedance of the insulation fault point. The greater the insulation degradation, the greater the insulation damage caused by the arc energy. Record the zero-mode voltage effective value U 0rms, expressed as the following formula,

[0146]

[0147] Where:

[0148] u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system;

[0149] N is the fault duration T D The number of sampling points within .

[0150] Fault interval time T DS : The time interval between this ground fault and the previous ground fault;

[0151] Number of instantaneous ground faults within a short time limit N ST : The number of instantaneous ground faults detected within a short time limit. The short time limit can be set to 1 minute, and the number of instantaneous ground faults can be set to 3;

[0152] Long-term instantaneous ground fault fault N LT : The number of instantaneous ground faults detected within a long time limit;

[0153] Number of ground faults within 24 hours N tD : The number of insulation failures within 24 hours, which indirectly reflects that insulation degradation has a tendency to develop into permanent breakdown in a short period of time;

[0154] Fault impact elimination time T CL : Set the maximum impact time T max , taking the i-th fault as an example, as time goes by, its contribution to the insulation degradation value should gradually decrease. It is set according to the maintenance cycle and fault elimination cycle. After the cable is eliminated or the insulation performance changes, the fault impact of the historical data should be solidified and calculated from the new fault data. It is expressed as the following formula:

[0155] T CL =T max -T ki

[0156] Where:

[0157] T max The maximum impact time of the fault is set according to the maintenance cycle, and is preferably set to 360 days;

[0158] T ki is the time interval between the current time and the i-th ground fault;

[0159] Permanent ground fault flag F PG : Ground fault duration T DIf it is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1;

[0160] Eliminate the number of faults within the time N fY : Fault influence elimination time T CL Number of instantaneous faults occurring within

[0161] Monthly failure rate change rate ΔV fm : Change rate of the number of failures in a natural month compared with the previous month.

[0162] Step S4: Define the insulation degradation evaluation index D i ∈[0, 1] evaluates the abnormality of cable insulation, divides the evaluation level, selects the evaluation factors from the valid data of the instantaneous grounding fault calculated by S3, determines the evaluation level of the instantaneous grounding fault through fuzzy calculation and issues the corresponding alarm signal.

[0163] In a preferred but non-limiting embodiment of the present invention, step S4 specifically includes:

[0164] Step S401: The insulation degradation evaluation index D is set according to the set threshold. i The size range of is divided into N continuous evaluation levels, representing different degrees of alarm; the threshold of the evaluation level can be adjusted according to the on-site operation conditions.

[0165] Further preferably, according to D i The inference calculation value is divided into four evaluation levels: emergency alarm [0.85, 1], serious alarm [0.75, 0.85], general alarm [0.5, 0.75], latent fault [0, 0.5), where D i ≥0.5 means that the cable has insulation abnormality, D i <0.5 means that there is a latent fault in the cable.

[0166] Step S402: Select the fault duration T from the valid data of the instantaneous ground fault. D , zero mode voltage effective value U 0rms And the fault interval time T DS As evaluation factors, trapezoidal membership functions are used for fuzzy processing, as follows:

[0167] The duration of instantaneous ground fault T D (Unit: ms) The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0168]

[0169] Where:

[0170] TD ' represents T after gelatinization D .

[0171] The effective value of zero-mode voltage U 0rms (Unit: V) The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0172]

[0173] Where:

[0174] U 0nms ' represents U after gelatinization 0rms .

[0175] The time interval between consecutive instantaneous faults is T DS (Unit: day) The trapezoidal membership function is used for fuzzy processing, which is expressed as follows:

[0176]

[0177] Where:

[0178] T DS ' represents T after gelatinization DS .

[0179] It is worth noting that other data or data combinations can also be selected from the instantaneous fault valid data as evaluation factors, and fuzzy processing can be performed using a set membership function. Based on the spirit of the present invention, all of these fall within the protection scope of the present invention.

[0180] Step S403, calculating the membership degree r of each evaluation factor to each evaluation level after fuzzy processing ij , with membership degree r ij The fuzzy relationship matrix R is established as matrix elements.

[0181] Further preferably, the membership is calculated based on the ascending semi-trapezoidal function, which is expressed by the following formula:

[0182]

[0183] Where:

[0184] r ij represents the membership of the i-th factor to the j-th evaluation level;

[0185] S j , S j+1 Represents two adjacent thresholds of the j-th evaluation level, for example, any two adjacent values ​​among 0, 0.5, 0.75, 0.85, and 1;

[0186] C iRepresents the calculated value of the membership function of the i-th factor.

[0187] Calculate r ij , we can get the fuzzy relationship matrix R, which can be expressed as follows:

[0188] R=[r ij ] M*N

[0189] Where:

[0190] M is the number of evaluation factors. When the fault duration T is D , zero mode voltage effective value U 0 And the continuous instantaneous fault interval time T DS When calculating the membership degree, M = 3;

[0191] N is the number of evaluation levels. When the evaluation levels are divided from high to low according to emergency alarm, serious alarm, general alarm and latent fault, N=4.

[0192] Step S404, using the weight matrix W and the fuzzy relationship matrix R to calculate the node insulation degradation degree evaluation index D i For the membership matrix B of each evaluation level, the insulation degradation degree evaluation index D is obtained according to the maximum membership principle. i If the membership matrix B has multiple maximum elements with the same value, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements.

[0193] Further preferably, the fault duration T is allocated D , zero mode voltage effective value U 0rms And the fault interval time T D The weights of the three evaluation factors are 0.3, 0.4, and 0.3 respectively. The sum of the weights of all evaluation factors is 1. The weight matrix is ​​constructed as W = [0.3, 0.4, 0.3], then B = W × R.

[0194] It can be understood that B is a 1×N matrix, and the element b in B 1j Representative node insulation degradation evaluation index D i The degree of membership to the ,th evaluation level.

[0195] Step S405, set D i The inference calculation supplementary rules for D i Adjust and determine the final insulation degradation evaluation index D i Value, the inference calculation supplementary rules are as follows:

[0196] 1) If the number of instantaneous ground faults within 24 hours is N tD If it is greater than 24 times, output 1;

[0197] 2) If the continuous instantaneous fault interval T in the historical fault data of the last month is DS If the difference is decreasing, output 1;

[0198] 3) If the most recent fault affects the elimination time T CL If it is greater than 30 days, +0.2 is added to the evaluation value result of step S404.

[0199] Step S406: Each node is evaluated based on the final insulation degradation degree evaluation index D i , determine the evaluation level of the instantaneous ground fault and output the corresponding alarm signal.

[0200] In order to facilitate the understanding of step S4, a typical case is selected as an example for illustration below.

[0201] When T D =5,U 0 =0, T DS =0, the fuzzy processing results are 0.5, 0, 0 respectively.

[0202] Then B=W*R=[0.3000], MAX[B] should output a latent warning.

[0203] When T D =7.5, U 0 =75, T DS =22.5, the fuzzy processing results are 0.75, 0.75, 0.75 respectively.

[0204] Then B=W*R=

[1100] , MAX[B] should output a general warning (latent and general conditions are met at the same time, and the output is based on the high-level evaluation).

[0205] When T D =0,U 0 =0, T DS = 0, the fuzzy processing results are 0, 0, and 1 respectively.

[0206] Then B=W*R=[0.30.30.30.3], MAX[B] should output an emergency alarm (latent and emergency are met at the same time, and the output is based on the higher level evaluation).

[0207] Step S5: Select different communication methods to transmit alarm signals and recording files according to whether lateral communication conditions are available, and preliminarily locate the cable insulation abnormality section or potential fault section.

[0208] Specifically, if there are optical fiber / 5G / 4G horizontal communication conditions between each node, the adjacent nodes share the alarm information and send the fault recording to the distribution automation master station at the same time. If a node and its upstream node both report a cable insulation abnormality alarm, and its downstream node does not alarm, the cable insulation abnormality section is preliminarily determined to be between the node and its downstream node; if there are only optical fiber / 5G / 4G and distribution automation master station vertical communication conditions between the nodes, the alarm information and fault recording of the node are sent up, and the distribution automation master station judges the alarm information sent by each node based on the topological structure of the distribution network, and preliminarily locates the cable insulation abnormality or potential fault section.

[0209] Step S6: Based on the topological structure of the distribution network, waveform comparison and analysis are performed on the recording files of adjacent nodes in the initially determined cable insulation abnormality or potential fault section, and at the same time, at least one node at the upper and lower levels of the preferred section is expanded for verification. If there is no error, a cable insulation abnormality or potential fault warning is issued, otherwise, it is classified as a system disturbance.

[0210] In a preferred but non-limiting embodiment of the present invention, step S6 specifically includes:

[0211] Step S601, select reference fault recording trigger time t 0 The triggering time of other fault recordings is t i , using the wide area clock synchronization, the sampling points of other recorded data are shifted. The specific number of sampling point shifts is n = |t i -t 0 |*f, align the fault recording data in the same area and compare the similarity of zero-sequence current waveforms of adjacent nodes after alignment;

[0212] Step S602, first select the transient recording of the extended node to perform zero-sequence current waveform similarity verification. If the transient recording waveform of the extended node fails to cover the waveform at the initial moment of the fault, then use the continuous recording file of the minute when the current transient recording file is triggered to verify the zero-sequence current waveform similarity.

[0213] Further preferably, if the fault is subsequently inverted, the zero-sequence current waveform similarity is checked in combination with the dynamic recording file.

[0214] Step S603, judging the latent fault or system disturbance according to the similarity of the zero-sequence current waveforms of the adjacent nodes and the extended nodes.

[0215] like Figure 2As shown, embodiment 2 of the present invention provides a cable insulation anomaly and latent fault early warning device based on electrical quantity, including the functions of three remote controls and protection of the distribution automation station terminal, on this basis, the instantaneous grounding fault identification and statistics of the line based on electrical quantity information are realized, and also includes: a real-time core, a management core, a communication module, a measurement and control module, an input and output module and a protection module;

[0216] The real-time core includes a ground fault determination unit, a wave recording and index data calculation unit;

[0217] The ground fault determination unit is used to extract transient fault information from the characteristic frequency bands of zero-sequence voltage and zero-sequence current to complete ground fault determination;

[0218] The wave recording and index data calculation unit is used to start the fault or disturbance analog data cache, generate a wave recording file, record and calculate the effective data of the instantaneous ground fault, and determine whether the number of instantaneous ground faults within a preset time limit exceeds a threshold value corresponding to the preset time limit;

[0219] The measurement and control module is used to synchronously monitor the zero-sequence voltage and zero-sequence current of the power distribution network system in real time;

[0220] The binary input and output module is used to collect and output binary quantities;

[0221] The protection module is used to determine whether the zero-sequence voltage or zero-sequence current meets the fault start condition and issue an alarm signal;

[0222] The management core includes a historical fault data storage unit, an insulation degradation index calculation and statistical data analysis unit, and a device status monitoring unit;

[0223] The historical fault data storage module is used to store the recording files and calculation results of the recording and index data calculation unit;

[0224] The insulation degradation index calculation and statistical data analysis unit is used to define the insulation degradation degree evaluation index and divide the evaluation level, and determine the evaluation level to which the instantaneous ground fault belongs based on the valid data of the instantaneous ground fault in the historical fault data storage module;

[0225] The device status monitoring unit is used to monitor the operation of the main body, the channel link status and the data abnormality alarm;

[0226] The communication module is used to transmit alarm information and fault recording according to communication conditions;

[0227] The real-time core and the management core operate as dual cores, and data interaction is achieved through the data bus;

[0228] The device also includes: a power-off management core, which is used to transfer and store data when the device is powered off; and a power-on real-time core, which is used to read the most recently transferred and stored data.

[0229] Preferably, the fault start condition in the protection module is the amplitude U of the collected system zero-sequence voltage 0 Exceeds the preset threshold U set , or the amplitude change of zero-sequence voltage ΔU 0 Exceeds the preset threshold ΔU set , or the amplitude of the system zero-sequence current I 0max Exceeds the preset threshold I set , or the amplitude change of the node zero-sequence current ΔI 0max Exceeds the preset threshold ΔI 0set .

[0230] Preferably, the preset time limit in the wave recording and index data calculation unit includes a short time limit or a long time limit, the short time limit is set to 1 minute, and the threshold corresponding to the short time limit is 3 times; the long time limit is set to 3 minutes, and the threshold corresponding to the long time limit is 5 times;

[0231] The valid data of instantaneous ground fault in the waveform recording and index data calculation unit include the fault duration T D , zero mode voltage effective value U 0rms , Fault interval time T DS , Number of instantaneous ground faults within a short time limit N ST , Long-term instantaneous ground fault fault N LT 、Number of ground faults within 24 hours tD , Fault impact elimination time T CL , Permanent ground fault sign F PG , the number of faults within the elimination time N fY , Monthly failure rate change rate ΔV fm ;

[0232] The recording and index data calculation unit records and calculates the effective data of the instantaneous ground fault, specifically including:

[0233] Fault duration T D : Taking the sampling value u of the zero-sequence voltage of the distribution network system 0 Exceeding the threshold value U for the first time set The starting time is less than U set And it does not exceed U again in the next power frequency cycle set The moment of is the termination moment;

[0234] Zero mode voltage effective value U 0rms : Record the zero-mode voltage effective value U 0rms , expressed as the following formula,

[0235]

[0236] Where:

[0237] u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system;

[0238] N is the fault duration T D The number of sampling points within

[0239] Fault interval time T DS : The time interval between this ground fault and the previous ground fault;

[0240] Number of instantaneous ground faults within a short time limit N ST : The number of instantaneous ground faults detected within a short time limit;

[0241] Long-term instantaneous ground fault fault N LT : The number of instantaneous ground faults detected within a long time limit;

[0242] Number of ground faults within 24 hours N tD : The number of insulation failures within 24 hours;

[0243] Fault impact elimination time T CL : Fault influence elimination time T CL The calculation formula is,

[0244] T CL =T max -T ki

[0245] Where:

[0246] T max The maximum impact time of the fault is set according to the maintenance cycle;

[0247] T ki is the time interval between the current time and the i-th ground fault;

[0248] Permanent ground fault flag F PG : Ground fault duration T D If it is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1;

[0249] Eliminate the number of faults within the time N fY : Fault influence elimination time T CL Number of instantaneous faults occurring within

[0250] Monthly failure rate change rate ΔV fm : Change rate of the number of failures in a natural month compared with the previous month.

[0251] Preferably, in the insulation degradation index calculation and statistical data analysis unit, determining the evaluation level to which the instantaneous ground fault belongs specifically includes:

[0252] According to the set threshold, the insulation degradation evaluation index D i The size range is divided into N continuous evaluation levels; M valid data are selected from the valid data of instantaneous grounding fault as evaluation factors, and fuzzy processing is performed using membership functions respectively; the membership r of each evaluation factor to each evaluation level after fuzzy processing is calculated ij , with membership degree r ij As matrix elements, the fuzzy relationship matrix R is established; the node insulation degradation degree evaluation index D is calculated using the weight matrix W and the fuzzy relationship matrix R i For the membership matrix B of each evaluation level, the insulation degradation degree evaluation index D is obtained according to the maximum membership principle. i If there are multiple maximum elements with the same value in the membership matrix B, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements; set D i The inference calculation supplementary rules for D i Adjust and determine the final insulation degradation evaluation index D i Value; Each node is evaluated according to the final insulation degradation degree index D i , determine the evaluation level of the instantaneous ground fault.

[0253] Preferably, the communication module shares alarm information with adjacent devices via a horizontal communication channel, or transmits alarm information and fault recording to a distribution automation master station via a vertical communication channel.

[0254] Embodiment 3 of the present invention provides a cable insulation anomaly and latent fault warning system based on electrical quantities, comprising: a distribution automation master station and multiple devices as described in Embodiment 2; the devices and the devices and the distribution automation master station are interconnected based on horizontal communication channels or vertical communication channels to form a system for segmented positioning and warning of insulation anomalies and latent faults of distribution network cables.

[0255] In a specific embodiment of the present invention, Figure 3 As shown, each node in the distribution network is installed with a cable insulation anomaly and potential fault warning device based on electrical quantity information as described in Example 2. Adjacent terminals share information horizontally through optical fibers, and the terminals and the distribution automation master station interact with each other vertically through 4G / 5G, forming a cable insulation anomaly and potential fault warning system based on electrical quantities that has both vertical and horizontal communications.

[0256] Taking the insulation abnormality of the F1 section cable as an example, the S1 switch and the ring network cabinet 1, 2 terminals detect that the system voltage exceeds the limit, start the fault or disturbance analog data cache, and generate a recording file at the same time; use Hilbert transform to perform time-frequency domain processing on the zero-mode voltage and current signals, and extract the transient fault information of the characteristic frequency band; the terminal performs fault analysis and judgment. If a grounding fault is determined on-site, the fault type is classified. In the case of a permanent grounding fault, a permanent grounding fault alarm signal is directly issued, otherwise it is classified as a transient grounding fault, and the transient fault valid data and insulation degradation degree evaluation index are recorded and calculated; if the number of transient grounding faults within a short time limit or a long time limit exceeds the threshold, the S1 switch and the ring network cabinet 1, 2 terminals share alarm information and send fault recordings to the distribution automation master station at the same time. The S1 terminal combines its own downstream alarm and the upstream alarm or no alarm signal of the ring main unit 1 to preliminarily determine that the abnormal cable insulation section is located at F1. Otherwise, if the insulation degradation index of the S1 switch and the ring main units 1 and 2 terminals is greater than the threshold, the alarm signal is shared. The S1 terminal combines its own insulation degradation alarm and the no alarm signal of the ring main unit 1 to preliminarily determine that the abnormal cable insulation section is located at F1. After the distribution automation master station aligns the recording data of the terminal devices of each node, it expands one level downstream to the ring main unit 2 according to F1, and uses the similarity of the zero-sequence current waveforms of adjacent nodes to check the abnormal cable insulation section. After confirming that it is correct, a latent fault warning will be issued.

[0257] In a specific embodiment of the present invention, Figure 4 As shown, each node in the distribution network is installed with a cable insulation anomaly and potential fault warning device based on electrical quantity information as described in Example 2, and vertical information exchange is carried out between the terminal and the distribution automation master station through 4G / 5G, forming a cable insulation anomaly and potential fault warning system based on electrical quantity with only vertical communication.

[0258] Taking the insulation abnormality of the F1 section cable as an example, the outgoing line switch CB, the section / branch switch FS1-FS3 and the boundary switch YS switch detect that the system voltage exceeds the limit, start the fault or disturbance analog data cache, and generate a recording file at the same time; use Hilbert transform to process the zero-mode voltage and current signals in the time and frequency domain to extract the transient fault information of the characteristic frequency band; the terminal performs fault analysis and judgment. If a grounding fault is determined on the spot, the fault type is classified. In the case of a permanent grounding fault, a permanent grounding fault alarm signal is directly issued, otherwise it is classified as a transient grounding fault, and the effective data of the transient fault and the evaluation index of the insulation degradation degree are recorded and calculated; if a transient grounding fault occurs within a short time limit or a long time limit, the terminal will If the number of faults or the evaluation index of insulation degradation degree exceeds the threshold, the alarm signal and fault recording of the distribution automation master station are sent to the distribution automation master station. Based on the topological information, the distribution automation master station finds that there are alarm signals on the outgoing line switch CB and the section / branch switch FS1, but no alarm signals on FS2-FS3, and locates the abnormal cable insulation between the section / branch switches FS1 and FS2. After aligning the recording data of the terminal devices of each node, the distribution automation master station expands one level upstream to the outgoing line switch CB and one level downstream to the section / branch switch FS3 and the boundary switch YS according to F1, and uses the similarity of the zero-sequence current waveforms of adjacent nodes to check the abnormal cable insulation section. After confirming that it is correct, a latent fault warning is issued.

[0259] Compared with the prior art, the beneficial effects of the present invention include at least:

[0260] (1) The present invention uses the electrical quantity information of each node in the distribution network system to realize the early warning of cable insulation abnormality and latent faults, without the need to add other equipment including sensors;

[0261] (2) The present invention evaluates the insulation status according to the frequency, duration, voltage amplitude, etc. of transient faults, and performs segment positioning based on network topology through field node information sharing or distribution automation master station. It effectively solves the problems of difficult insulation status monitoring of distribution network cables, post-fault handling of multiple faults, and unplanned power outages that reduce power supply reliability, and has broad practical application value.

[0262] The present disclosure may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present disclosure.

[0263] Finally, 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 above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A cable insulation anomaly and latent fault early warning method based on electrical quantities, characterized in that: The following steps are involved: Step S1: Real-time synchronous monitoring of the zero-sequence voltage and zero-sequence current of the distribution network system. When the zero-sequence voltage or zero-sequence current meets the fault start condition, the fault or disturbance analog data cache is started and a recording file is generated at the same time. Otherwise, the zero-sequence voltage and zero-sequence current of the distribution network system are continuously monitored. Step S2: extracting transient fault information from the characteristic frequency bands of the zero-sequence voltage and the zero-sequence current to complete ground fault identification. If no ground fault is identified, uploading the waveform recording file for verification and ending the step. If a ground fault is identified, determining whether the fault is a permanent fault based on the transient fault information. If it is a permanent fault, directly issuing an alarm signal and ending the step. Otherwise, it is classified as a transient ground fault. Step S3: Record and calculate the valid data of instantaneous ground faults, and determine whether the number of instantaneous ground faults within a preset time limit exceeds the threshold corresponding to the preset time limit. If so, it is determined that the insulation of the cable in the downstream area of ​​this node is abnormal, an alarm signal is issued and the process jumps to step S5; Step S4: defining the insulation degradation evaluation index and dividing the evaluation levels, and based on the valid data of the instantaneous ground fault in S3, determining the evaluation level to which the instantaneous ground fault belongs and issuing a corresponding alarm signal; Step S5: transmitting alarm signals and recording files according to communication conditions, and preliminarily locating the cable insulation abnormality section or potential fault section; Step S6: Compare the waveforms in the recording files of the adjacent nodes of the initially determined cable insulation abnormality or latent fault section, expand the insulation abnormality or latent fault section nodes for verification, and issue a cable insulation abnormality or latent fault warning if it is correct, otherwise, it is classified as a system disturbance.

2. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: The fault initiation condition is that the amplitude U0 of the collected system zero-sequence voltage exceeds the preset threshold U set , or the amplitude change of the zero-sequence voltage ΔU0 exceeds the preset threshold ΔU set , or the amplitude of the system zero-sequence current I 0max Exceeds the preset threshold I set , or the amplitude change of the node zero-sequence current ΔI 0max Exceeds the preset threshold ΔI 0set .

3. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S1, Beidou, GPS, IRIG-B code or SNTP is used to synchronize the time of the synchronous monitoring process.

4. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S1, the types of recording files include transient recording, continuous recording and dynamic recording; the recording files adopt the standard COMTRADE format, the sampling rate f of transient recording is not less than 12.8kHz, and at least contains data 80ms before and 100ms after the trigger moment; continuous recording forms a file per minute, and the sampling rate is not less than 1.6kHz; the dynamic recording rate is not less than 25 times / second.

5. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S2, Hilbert transform is used to perform time-frequency domain processing on the zero-sequence voltage and zero-sequence current to extract transient fault information in a characteristic frequency band.

6. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S2, the lower limit of the characteristic frequency band is 150 Hz and the upper limit is 2000 Hz.

7. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S2, whether there is a ground fault is determined locally by any one of the following methods: a DC offset method, a transient power direction method, a first half-wave method, and a three-phase current imbalance method.

8. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1 is characterized by: In step S3, the preset time limit includes a short time limit or a long time limit. The short time limit is set to 1 minute, and the threshold corresponding to the short time limit is 3 times; the long time limit is set to 3 minutes, and the threshold corresponding to the long time limit is 5 times.

9. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 8, characterized in that: The valid data of instantaneous ground fault include fault duration T D , zero mode voltage effective value U 0rms , Fault interval time T DS , Number of instantaneous ground faults within a short time limit N ST , Long-term instantaneous ground fault fault N LT 、Number of ground faults within 24 hours tD , Fault impact elimination time T CL , Permanent ground fault sign F PG , the number of faults within the elimination time N fY , Monthly failure rate change rate ΔV fm ; In step S3, recording and calculating the effective data of the instantaneous ground fault specifically includes: Fault duration T D :The sampling value u0 of the zero-sequence voltage of the distribution network system exceeds the threshold value U for the first time set The starting time is less than U set And it does not exceed U again in the next power frequency cycle set The moment of is the termination moment; Zero mode voltage effective value U 0rms : Record the zero-mode voltage effective value U 0rms , expressed as the following formula, Where: u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system; N is the fault duration T D The number of sampling points within Fault interval time T DS : The time interval between this ground fault and the previous ground fault; Number of instantaneous ground faults within a short time limit N ST : The number of instantaneous ground faults detected within a short time limit; Long-term instantaneous ground fault fault N LT : The number of instantaneous ground faults detected within a long time limit; Number of ground faults within 24 hours N tD : The number of insulation failures within 24 hours; Fault impact elimination time T CL : Fault influence elimination time T CL The calculation formula is, T CL =T max -T ki Where: T max The maximum impact time of the fault is set according to the maintenance cycle; T ki is the time interval between the current time and the i-th ground fault; Permanent ground fault flag F PG : Ground fault duration T D If it is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1; Eliminate the number of faults within the time N fY : Fault influence elimination time T CL Number of instantaneous faults occurring within Monthly failure rate change rate ΔV fm : Change rate of the number of failures in a natural month compared with the previous month.

10. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 1, characterized in that: Step S4 specifically includes: Step S401: The insulation degradation evaluation index D is set according to the set threshold. i The size range of is divided into N continuous evaluation levels; Step S402, selecting M valid data from the valid data of the instantaneous ground fault as evaluation factors, and performing fuzzy processing on each evaluation factor using a membership function; Step S403, calculating the membership degree r of each evaluation factor to each evaluation level after fuzzy processing ij , with membership degree r ij As matrix elements, establish the fuzzy relationship matrix R; Step S404, using the weight matrix W and the fuzzy relationship matrix R to calculate the node insulation degradation degree evaluation index D i For the membership matrix B of each evaluation level, the insulation degradation degree evaluation index D is obtained according to the maximum membership principle. i If there are multiple maximum elements with the same value in the membership matrix B, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements; Step S405, set D i The inference calculation supplementary rules for D i Adjust and determine the final insulation degradation evaluation index D i value; Step S406: Each node is evaluated based on the final insulation degradation degree evaluation index D i , determine the evaluation level of the instantaneous ground fault and output the corresponding alarm signal.

11. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 10, characterized in that: Setting D i The size range is [0,1], and each threshold is set, including 0.5, 0.75, and 0.

85. According to each threshold, the insulation degradation evaluation index D i It is divided into four evaluation levels, including emergency alarm, severe alarm, general alarm, and latent fault; the size range of emergency alarm is [0.85, 1], the size range of severe alarm is [0.75, 0.85), the size range of general alarm is [0.5, 0.75), and the size range of latent fault is [0, 0.5).

12. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 10, characterized in that: Select the fault duration T D , zero mode voltage effective value U 0rms And the fault interval time T DS As an evaluation factor, the instantaneous ground fault duration T D The trapezoidal membership function is used for fuzzy processing, which is expressed as follows: Where: T D ' represents T after gelatinization D ; The effective value of zero-mode voltage U 0rms The trapezoidal membership function is used for fuzzy processing, which is expressed as follows: Where: U 0rms ′ U after gelatinization 0rms ; The time interval between consecutive instantaneous faults is T DS The trapezoidal membership function is used for fuzzy processing, which is expressed as follows: Where: T DS ' represents T after gelatinization DS .

13. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 10, characterized in that: In step S403, the degree of membership is calculated based on the descending semi-trapezoidal function, which is expressed as follows: Where: r ij represents the membership of the i-th evaluation factor to the j-th evaluation level; S j , S j+1 Represents two adjacent thresholds of the jth evaluation level; C i Represents the calculated value of the membership function of the i-th evaluation factor; Calculate r ij , we get the fuzzy relationship matrix R, which is expressed as follows: R＝[r ij ] M*N Where: M is the number of evaluation factors; N is the number of evaluation levels.

14. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 12, characterized in that: In step S404, the fault duration T is allocated D , zero mode voltage effective value U 0rms And the fault interval time T DS The weight of the fault duration T D , zero mode voltage effective value U 0rms And the fault interval time T DS The weights of are used to construct the weight matrix W, then B = W × R.

15. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 10, characterized in that: In step S405, three inference calculation supplementary rules are set, and the three inference calculation supplementary rules are: 1) If the number of instantaneous ground faults within 24 hours is N tD If it is greater than 24 times, output 1; 2) If the continuous instantaneous fault interval T in the historical fault data of the last month is DS If the difference is decreasing, output 1; 3) If the most recent fault affects the elimination time T CL If it is greater than 30 days, +0.2 is added to the evaluation value result of step S404.

16. The cable insulation anomaly and latent fault early warning method based on electrical quantities according to claim 4, characterized in that: Step S6 specifically includes: Step S601, select the reference fault recording trigger time t0, and the other fault recording trigger times are t i ,Using the wide area clock synchronization, the sampling points of other recorded data are shifted, the fault recorded data in the same area are aligned, and the zero-sequence current waveforms of adjacent nodes after alignment are compared; Step S602, perform zero-sequence current waveform similarity check on the transient recording of the extended node. If the transient recording waveform of the extended node fails to cover the waveform at the initial moment of the fault, then use the continuous recording file of the minute at which the current transient recording file is triggered to check the zero-sequence current waveform similarity. If the fault is subsequently inverted, then combine the dynamic recording file to check the zero-sequence current waveform similarity. Step S603, judging the latent fault or system disturbance according to the similarity of the zero-sequence current waveforms of the adjacent nodes and the extended nodes.

17. A cable insulation anomaly and latent fault early warning device based on electrical quantities, with three remote control and protection functions of distribution automation station terminals, characterized in that: include: Real-time core, management core, communication module, measurement and control module, input and output module and protection module; The real-time core includes a ground fault determination unit, a wave recording and index data calculation unit; The ground fault determination unit is used to extract transient fault information from the characteristic frequency bands of zero-sequence voltage and zero-sequence current to complete ground fault determination; The wave recording and index data calculation unit is used to start the fault or disturbance analog data cache, generate a wave recording file, record and calculate the effective data of the instantaneous ground fault, and determine whether the number of instantaneous ground faults within a preset time limit exceeds a threshold value corresponding to the preset time limit; The measurement and control module is used to synchronously monitor the zero-sequence voltage and zero-sequence current of the power distribution network system in real time; The binary input and output module is used to collect and output binary quantities; The protection module is used to determine whether the zero-sequence voltage or zero-sequence current meets the fault start condition and issue an alarm signal; The management core includes a historical fault data storage unit, an insulation degradation index calculation and statistical data analysis unit, and a device status monitoring unit; The historical fault data storage module is used to store the recording files and calculation results of the recording and index data calculation unit; The insulation degradation index calculation and statistical data analysis unit is used to define the insulation degradation degree evaluation index and divide the evaluation level, and determine the evaluation level to which the instantaneous ground fault belongs based on the valid data of the instantaneous ground fault in the historical fault data storage module; The device status monitoring unit is used to monitor the operation of the main body, the channel link status and the data abnormality alarm; The communication module is used to transmit alarm information and fault recording according to communication conditions; The real-time core and the management core operate as dual cores, and data interaction is achieved through the data bus; The device also includes: a power-off management core for transferring and storing data when the device loses power; The real-time core is used to read the most recently transferred and stored data.

18. The cable insulation anomaly and latent fault early warning device based on electrical quantities according to claim 17, characterized in that: The fault start condition in the protection module is that the amplitude U0 of the collected system zero-sequence voltage exceeds the preset threshold U set , or the amplitude change of the zero-sequence voltage ΔU0 exceeds the preset threshold ΔU set , or the amplitude of the system zero-sequence current I 0max Exceeds the preset threshold I set , or the amplitude change of the node zero-sequence current ΔI 0max Exceeds the preset threshold ΔI 0set .

19. The cable insulation anomaly and latent fault early warning device based on electrical quantities according to claim 17, characterized in that: The preset time limit in the waveform recording and index data calculation unit includes a short time limit or a long time limit. The short time limit is set to 1 minute, and the threshold corresponding to the short time limit is 3 times; the long time limit is set to 3 minutes, and the threshold corresponding to the long time limit is 5 times; The effective data of instantaneous ground fault in the waveform recording and index data calculation unit include fault duration TD, zero-mode voltage effective value U 0rms , fault interval time TDS, number of instantaneous ground faults within a short time limit NST, number of instantaneous ground faults within a long time limit NLT, number of ground faults within 24 hours NtD, fault impact elimination time TCL, permanent ground fault flag FPG, number of faults within the elimination time NfY, monthly fault number change rate ΔVfm; The recording and index data calculation unit records and calculates the effective data of the instantaneous ground fault, specifically including: Fault duration TD: The starting time is the time when the sampling value u0 of the zero-sequence voltage of the distribution network system exceeds the threshold value Uset for the first time, and the ending time is the time when it is less than Uset and does not exceed Uset again within the next power frequency cycle; Zero mode voltage effective value U 0rms : Record the zero-mode voltage effective value U 0rms , expressed as the following formula, Where: u 0j Represents the jth sampling value of the zero-sequence voltage of the distribution network system; N is the number of sampling points within the fault duration TD; Fault interval time TDS: the time interval between the current ground fault and the previous ground fault; Number of instantaneous ground faults within a short time limit NST: The number of instantaneous ground faults detected within a short time limit; Long-term transient ground fault faults NLT: The number of transient ground faults detected within a long time period; Number of ground faults within 24 hours NtD: Number of insulation faults within 24 hours; Fault impact elimination time TCL: The calculation formula for fault impact elimination time TCL is, TCL=Tmax-Tki Where: Tmax is the maximum impact time of the fault, which is set according to the maintenance cycle; Tki is the time interval between the current time and the ith ground fault; Permanent ground fault flag FPG: If the ground fault duration TD is greater than the judgment time, it is considered that the instantaneous ground fault is converted to a permanent ground fault, and the flag is set to 1; Number of faults within the elimination time NfY: the number of instantaneous faults that occur within the fault impact elimination time TCL; Monthly failure rate change rate ΔVfm: The change rate of the failure rate between the natural month and the previous month.

20. The cable insulation anomaly and latent fault early warning device based on electrical quantities according to claim 17, characterized in that: In the insulation degradation index calculation and statistical data analysis unit, determining the evaluation level of the instantaneous ground fault specifically includes: According to the set threshold, the size range of the insulation degradation degree evaluation index Di is divided into N continuous evaluation levels; M valid data are selected from the valid data of the instantaneous grounding fault as evaluation factors, and fuzzy processing is performed respectively using the membership function; the membership rij of each evaluation factor to each evaluation level after fuzzy processing is calculated, and the fuzzy relationship matrix R is established with the membership rij as the matrix element; the membership matrix B of the node insulation degradation degree evaluation index Di to each evaluation level is calculated using the weight matrix W and the fuzzy relationship matrix R, and the numerical result of the insulation degradation degree evaluation index Di is obtained according to the maximum membership principle. If there are multiple maximum elements with the same value in the membership matrix B, the evaluation level is determined to be the highest evaluation level among the evaluation levels corresponding to the multiple maximum elements; the inference calculation supplementary rules of Di are set, Di is adjusted, and the final insulation degradation degree evaluation index Di value is determined; each node determines the evaluation level to which the instantaneous grounding fault belongs according to the final insulation degradation degree evaluation index Di.

21. The cable insulation anomaly and latent fault early warning device based on electrical quantities according to claim 17, characterized in that: The communication module shares alarm information with adjacent devices through a horizontal communication channel, or transmits alarm information and fault recording to a distribution automation master station through a vertical communication channel.

22. A cable insulation anomaly and latent fault early warning system based on electrical quantities, characterized in that: include: A distribution automation master station and a plurality of devices as described in any one of claims 17 to 21; The devices and the devices and the distribution automation master station are interconnected based on a horizontal communication channel or a vertical communication channel to form a system, which performs segmented positioning and early warning of insulation anomalies and latent faults of distribution network cables.

23. A computer device, characterized in that: include: one or more processors; The processor is used to store one or more programs; When the one or more programs are executed by the one or more processors, the cable insulation anomaly and latent fault early warning method based on electrical quantities as described in any one of claims 1 to 16 is implemented.

24. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and when the computer program is executed, the cable insulation anomaly and latent fault early warning method based on electrical quantities as described in any one of claims 1 to 16 is implemented.

Images