Line insulation hidden danger monitoring method and system based on multi-parameter analysis

Abstract

The present invention discloses a line insulation hidden danger monitoring method and system based on multi-parameter analysis, which belongs to the technical field of power grid hidden danger identification. According to the risk stress mechanism, the risk characteristic information of each busbar or / and branch line in the distribution network automation monitoring network is obtained in real time; the risk factor in the risk characteristic information is extracted, and the risk degree of the risk characteristic information is determined according to the risk factor combined with the risk value; the hidden danger point is determined according to the risk degree and its corresponding positioning strategy, and corresponding hidden danger disposal measures are formulated. The present application determines the fault point by analyzing multi-dimensional data, and determines the risk degree of the fault point by analyzing the degradation trend of the hidden danger, thereby realizing real-time accurate dynamic identification of distributed hidden dangers, and at the same time determines the corresponding inspection strategy according to the risk degree of the hidden danger and the maintenance priority, thereby improving the accuracy and efficiency of fault disposal and further reducing the further expansion of safety hazards.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid hidden danger identification, and in particular to a line insulation hidden danger monitoring method and system based on multi-parameter analysis. Background Art

[0002] In the power system, cable lines are important infrastructure for power transmission and distribution. Their safe and stable operation is of great significance to ensure the reliability and safety of the power grid. However, since cable lines are in a complex operating environment for a long time and are affected by various internal and external factors, such as temperature, humidity, electromagnetic interference, mechanical stress, etc., there are many potential safety hazards in cable lines. If these hidden dangers are not discovered and handled in time, they are very likely to cause cable failures and even cause large-scale power outages in the power grid, causing huge losses to social production and people's lives. In order to timely discover the negative effects of cable insulation degradation and discharge, in the field of cable operation and maintenance, traditional hidden danger identification methods mainly rely on manual inspections and regular inspections. However, this method has many shortcomings. First, manual inspections are easily affected by environmental and weather factors, resulting in the fact that the insulation hidden dangers of degradation may not necessarily have discharge phenomena at the time of detection. Secondly, some cable locations are hidden, which is not easy to detect by manual and partial discharge instruments, and there are blind spots. Thirdly, the manpower investment cost of periodic inspections is large, and it is unavoidable that the insulation state will further deteriorate during the interval between two inspections, leading to accidents. In addition, traditional hidden danger identification methods are often based on single-dimensional data analysis, and it is difficult to analyze the degradation trend of hidden danger points. They lack comprehensiveness and accuracy, and are prone to false alarms and omissions.

[0003] With the continuous development of distribution networks, the structure of cable networks has become more and more complex, and potential hazards have become distributed and diversified. Therefore, traditional potential hazard identification methods can no longer meet the needs of modern cable operation and maintenance. How to achieve real-time, accurate and dynamic identification of cable hazards, and formulate reasonable inspection strategies based on the risk level and maintenance priority of the hazards, has become an urgent problem to be solved in the current cable operation and maintenance field.

[0004] Chinese patent, invention patent with application number 2024100960921 "A method for detecting corona hazards of cable insulation degradation in distribution lines". The detection method described in this patent is specifically as follows: configure the terminal and master station required for corona hazard detection; the distribution automation equipment collects zero-sequence current, detects and uploads the instantaneous zero-sequence overcurrent in real time; retrieves the zero-sequence current sampling and recording data corresponding to the instantaneous zero-sequence overcurrent alarm signal for corona discharge waveform recognition; locates the corona discharge position based on the corona discharge waveform recognition result, and evaluates the risk value of the distribution line, so as to prioritize on-site live detection. Although this application can also locate the hidden danger position and inspect the hidden danger according to priority, the data dimension for judging the hidden danger is too single, and there is no analysis and judgment on the degradation trend of the hidden danger, which makes it difficult to determine the actual risk level of the hidden danger point, which in turn leads to the problem of unreliable maintenance strategy formulated according to the inspection priority.

[0005] The above information disclosed in the Background section is only for enhancement of understanding of the background of the present application and therefore it may contain information that does not constitute the prior art that is already known to a person of ordinary skill in the art. Summary of the invention

[0006] In view of the problem that it is difficult to dynamically monitor the degradation trend of hidden dangers in the above-mentioned background technology, resulting in unreliable inspection strategies, the present invention proposes a line insulation hidden danger monitoring method and system based on multi-parameter analysis. The fault point is determined by analyzing multi-dimensional data, and the degradation trend of the hidden danger is analyzed to determine the risk level of the fault point, thereby realizing real-time and accurate dynamic identification of distributed hidden dangers. At the same time, the corresponding inspection strategy is determined according to the risk level of the hidden danger and the maintenance priority, which improves the accuracy and efficiency of fault handling and further prevents the further expansion of safety hazards.

[0007] In a first aspect, a technical solution provided in an embodiment of the present invention is a line insulation hidden danger monitoring method based on multi-parameter analysis, comprising the following steps:

[0008] S1. Obtain risk characteristic information of each busbar and / or branch line in the distribution network automation monitoring network in real time according to the risk stress mechanism;

[0009] S2. extracting risk factors from the risk characteristic information, and determining the risk level of the risk characteristic information based on the risk factors and the risk value;

[0010] S3. Determine potential hazard points based on the risk level and their corresponding positioning strategies, and formulate corresponding potential hazard disposal measures.

[0011] In this scheme, firstly, the risk characteristic information of each busbar or / and branch line in the distribution network automation network is obtained in real time according to the risk stress mechanism. By collecting these multi-dimensional data, a comprehensive and accurate information basis is provided for the subsequent hidden danger identification; after obtaining the risk characteristic information, the present invention extracts the risk factors therein, and determines the risk degree of the risk characteristic information according to the risk factors combined with the risk value. Through the comprehensive analysis and calculation of the risk factors, the degree of safety hazards of the cable line can be more accurately evaluated. By comparing the risk degree, the risk characteristic information with hidden dangers is screened out, and the specific location of the hidden danger point is determined according to the source attribute information. At the same time, the degree of danger and degradation corresponding to the risk point is determined according to the network topology, such as the scope of hidden danger, maintenance time, waveform mutation trend and alarm trend, so as to formulate a scientific and reasonable maintenance priority and corresponding maintenance strategy, improve the accuracy and efficiency of fault handling, and further reduce the further expansion of safety hazards.

[0012] Preferably, in S1, the risk characteristic information of each busbar and / or branch line in the distribution network automation monitoring network is obtained in real time according to the risk stress mechanism; the steps include:

[0013] The risk stress mechanism is set according to the waveform characteristics of the zero-sequence current. When the risk stress mechanism is responded, the risk characteristic information of each busbar and / or branch line of the distribution network automation is obtained;

[0014] The risk characteristic information includes source attribute information, time attribute information and characteristic attribute information.

[0015] Preferably, the characteristic attribute information includes at least zero-sequence alarm information, recorded waveform information, voiceprint information and environmental information.

[0016] Preferably, in S2, the step of extracting risk factors from risk feature information comprises the following steps:

[0017] The alarm times and alarm trends in the zero-sequence alarm information of instantaneous return within the sampling time scale are obtained, and the alarm risk factor is determined according to the alarm trend, the alarm times and the normalization algorithm.

[0018] Preferably, in S2, extracting the risk factors from the risk feature information further comprises the following steps:

[0019] Get the corresponding record of the zero-sequence alarm with instantaneous reset within the sampling time scale (reset time is less than 40ms)

[0020] The number of zero-sequence spikes, the amplitude of zero-sequence spikes, the duration of zero-sequence spikes, the trend of spike mutations, and the comprehensive judgment results of the amplitude of zero-sequence spikes with the same mother at the same time in the waveform information;

[0021] The spike risk factor is determined based on the number of zero-sequence spikes, the amplitude of zero-sequence spikes, the duration of zero-sequence spikes, the amplitude of zero-sequence spikes with the same mother at the same time, and the spike mutation trend combined with the normalization algorithm;

[0022] Among them, the comprehensive analysis results of the zero-sequence spike amplitude of the same bus at the same time include: obtaining the instantaneous zero-sequence alarm set of all outgoing lines of the same bus within the sampling time scale and comparing the amplitudes of the alarm zero-sequence waveforms; determining the alarm interval corresponding to the line with the maximum amplitude of the zero-sequence waveform to increase the same-bus alarm weight.

[0023] Preferably, in S2, extracting the risk factors from the risk feature information further comprises the following steps:

[0024] Acquire the voiceprint feature information in the voiceprint information within the sampling time scale, and the time adaptability between the voiceprint mutation moment and the instantaneous zero-sequence current alarm moment;

[0025] The voiceprint risk factor is determined based on the voiceprint feature information, the adaptability of the voiceprint mutation and the instantaneous zero-sequence current alarm moment combined with the normalization algorithm.

[0026] Preferably, in S2, extracting the risk factors from the risk feature information further comprises the following steps:

[0027] Obtain the humidity change trend in the environmental information within the sampling time scale and the proportion of instantaneous zero-sequence alarms occurring in high humidity periods;

[0028] The temperature and humidity risk factors are determined based on the humidity change trend and the proportion of instantaneous zero-sequence alarms occurring at high humidity moments combined with a normalized algorithm.

[0029] Preferably, in S2, determining the risk level of the risk characteristic information according to the risk factor combined with the risk value includes the following steps:

[0030] The IV value algorithm is used to obtain the risk value of each characteristic attribute information relative to historical hidden dangers;

[0031] The risk value of risk characteristic information is determined by weighted summation of risk value and risk factor;

[0032] The degree of risk is determined by comparing the risk value with the risk level range.

[0033] Preferably, in S3, determining the potential danger point according to the risk degree and its corresponding positioning strategy, and formulating corresponding potential danger disposal measures, comprises the following steps:

[0034] S31. Compare and select risk characteristic information with hidden dangers based on risk levels;

[0035] S32, determining a potential danger point according to source attribute information corresponding to risk characteristic information of a potential danger; determining a danger level corresponding to the potential danger point according to a network topology;

[0036] S33, determining the maintenance priority according to the risk level and the corresponding critical situation level to obtain a corresponding maintenance strategy;

[0037] S34. The execution carrier performs maintenance actions in response to the maintenance strategy.

[0038] In a second aspect, a technical solution also provided in an embodiment of the present invention is a line insulation hidden danger monitoring system, comprising: a data acquisition unit: acquiring risk characteristic information of each busbar and / or branch line in a distribution network automation network in real time according to a risk stress mechanism;

[0039] Analysis unit: extract risk factors from risk feature information, and determine the risk level of risk feature information based on the risk factors and risk value;

[0040] Execution unit: Determine potential danger points based on the risk level and their corresponding positioning strategies, and formulate corresponding potential danger disposal measures.

[0041] In a third aspect, a technical solution provided in an embodiment of the present invention is an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the line insulation hidden danger monitoring method based on multi-parameter analysis are implemented.

[0042] In a fourth aspect, a technical solution provided in an embodiment of the present invention is a storage medium, in which computer executable instructions are stored. When the computer executable instructions are loaded and executed by a processor, the steps of the line insulation hazard monitoring method based on multi-parameter analysis are implemented.

[0043] Beneficial effects of the present invention:

[0044] (1) This application adopts a dynamic monitoring technology based on a risk stress mechanism, which can obtain risk characteristic information of each bus or branch in the distribution network automation network in real time. This mechanism ensures the timeliness and accuracy of the data, allowing the system to respond quickly to potential safety hazards. By collecting and analyzing multi-dimensional data in real time, the system can promptly detect abnormal changes in the cable line, providing strong support for subsequent hazard identification and disposal;

[0045] (2) This application achieves accurate dynamic identification of cable hazards by extracting multiple risk factors from risk feature information, such as alarm risk factors, spike risk factors, voiceprint risk factors, and temperature and humidity risk factors, and combining them with risk value for comprehensive evaluation. This multi-dimensional data analysis method fully considers various influencing factors in the operation of cable lines and improves the accuracy and reliability of hazard identification;

[0046] (3) This application provides a scientific basis for the formulation of inspection strategies through risk assessment. The system can automatically screen out risk feature information of hidden dangers according to the risk level and maintenance priority of hidden dangers, and determine the specific location and degradation trend of hidden danger points according to source attribute information. At the same time, combined with the network topology, the system can also evaluate the degree of danger corresponding to the hidden danger point, such as the scope of hidden dangers, recommended maintenance time, etc., so as to formulate scientific and reasonable maintenance priorities and corresponding maintenance strategies; this inspection strategy based on risk level not only improves the inspection efficiency, but also ensures the pertinence and effectiveness of the inspection work;

[0047] (4) This application has the ability to dynamically adjust and optimize. As the operating status of the power grid changes and the hidden danger identification technology continues to develop, the system can dynamically adjust and optimize the risk stress mechanism, risk factor extraction algorithm, and risk level assessment model according to actual conditions to ensure the accuracy and adaptability of the system in the face of complex power grid scenarios.

[0048] The above invention content is only an overview of the technical solution of the present invention. In order to more clearly understand the technical means of the present invention, it can be implemented according to the contents of the specification. In order to make the above and other purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Other features, objects and advantages of the present invention will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings. The drawings are only for the purpose of illustrating preferred embodiments and are not to be considered as limiting the present invention. Also, the same reference symbols are used throughout the drawings to represent the same parts.

[0050] Figure 1 This is a flow chart of the line insulation hidden danger monitoring method based on multi-parameter analysis of the present invention.

[0051] Figure 2 This is an exemplary distribution automation network topology diagram of the present invention.

[0052] Figure 3 This is a flow chart of the hidden danger location and disposal strategy of the present invention.

[0053] Figure 4It is a schematic diagram of the structure of the line insulation hidden danger monitoring system of the present invention. DETAILED DESCRIPTION

[0054] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific implementation method described herein is only an optimal embodiment of the present invention, which is only used to explain the present invention and does not limit the scope of protection of the present invention. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0055] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow charts describe the operations (or steps) as sequential processes, many of the operations (or steps) therein can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, etc.

[0056] Example 1: Figure 1 As shown, the line insulation hidden danger monitoring method based on multi-parameter analysis includes the following steps:

[0057] S1. Obtain risk characteristic information of each busbar and / or branch line in the distribution automation network in real time according to the risk stress mechanism.

[0058] Specifically, S1 includes the following steps:

[0059] A risk stress mechanism is set according to the waveform characteristics of the zero-sequence current. When the risk stress mechanism is responded to, risk characteristic information of each busbar and / or branch line of the distribution network automation is obtained;

[0060] The risk characteristic information includes source attribute information, time attribute information and characteristic attribute information.

[0061] In this embodiment, zero-sequence current sensors are deployed at the master station and each slave station (substation) to collect zero-sequence current data in real time; the collected zero-sequence current data is preprocessed, wherein the preprocessing steps mainly include filtering, denoising, etc. to ensure the accuracy of the data.

[0062] Waveform feature analysis: By analyzing the historical data that caused the fault, the features such as waveform distortion, spikes, frequency changes, etc. that are usually associated with abnormal conditions in the power grid (such as insulation degradation, corona discharge, etc.) are determined. Based on historical data and expert experience, this series of waveform feature thresholds are set to determine whether the current zero-sequence current waveform is abnormal. When the zero-sequence current waveform feature collected in real time exceeds the set threshold, the risk stress mechanism is triggered.

[0063] It is understandable that in a network topology consisting of a master station and multiple slave stations, the busbar, as the main channel for power transmission, is crucial to the stability of the entire power grid; while the branch line connects various loads, and its status may affect the normal operation of specific loads. After the risk stress mechanism is triggered, the specific location (bus or branch line) where risk characteristic information needs to be obtained is determined according to the network topology. When the master station detects that the waveform characteristics uploaded by the zero-sequence current sensor connected to one and only one bus ring network interval in a single distribution station are abnormal, the risk stress mechanism is triggered. At this time, the master station will request to obtain the risk characteristic information related to the bus, including the source attribute information of the bus and its directly connected branch line (determined to be derived from the bus), time attribute information and characteristic attribute information (such as zero-sequence alarm information, recorded waveform information, voiceprint information, environmental information, etc.). When the switch station detects that the waveform characteristics uploaded by the zero-sequence current sensor on the branch line connected to its bus are abnormal, the risk stress mechanism is triggered, and the slave station will immediately upload the risk characteristic information of the branch line to the master station. At the same time, the master station will request to obtain the bus risk characteristic information related to the branch line for comprehensive analysis and judgment.

[0064] A specific example is as follows, which is used to explain the above scheme. Figure 2 As shown, assuming a network topology that includes a substation outgoing line and three switch stations (stations A, B, and C):

[0065] Scenario 1: The master station detects that the waveform characteristics uploaded by the zero-sequence current sensor of the ring-in interval of the station B bus M are abnormal (such as the peak current exceeds the threshold), and at the same time, there is no instantaneous zero-sequence alarm and waveform abnormality in the ring-out interval of the station B bus M. At this time, the risk stress mechanism is triggered, and the master station is requested to obtain the risk characteristic information related to the station B bus M, including the instantaneous zero-sequence current alarm and amplitude of the ring-in and ring-out interval of the power supply side station A, whether there are instantaneous zero-sequence current alarms and amplitudes in the substation outgoing line intervals X, Y, and Z, and whether there are instantaneous zero-sequence current alarms and amplitudes in the incoming line intervals of the head-end switch station G and switch station D of other lines on the same bus. The largest amplitude comparison is the ring-in interval of the station B bus M, and the master station locates it as the risk characteristic information of the bus M;

[0066] Scenario 2: The waveform characteristics uploaded by the zero-sequence current sensor on the branch line L1 connected to station A and its bus M from the slave station monitoring are abnormal. At this time, the risk stress mechanism of slave station A is triggered, and it will confirm that the instantaneous zero-sequence alarm of the ring-in interval of station A and the waveform characteristics are abnormal, and at the same time, there is no instantaneous zero-sequence alarm in the ring-out interval of station A. If the conditions are met, the risk characteristic information of branch line L1 will be uploaded to the master station, and the master station will be requested to obtain the risk characteristic information related to the bus M of station A, including whether there are instantaneous zero-sequence current alarms and amplitudes in the outgoing line intervals X, Y, and Z of the substation, and whether there are instantaneous zero-sequence current alarms and amplitudes in the incoming line intervals of the head-end switch station G and switch station D of other lines with the same bus, so that the master station can conduct a comprehensive analysis and judgment on the hidden dangers of bus M and branch line L1.

[0067] Furthermore, the characteristic attribute information includes at least zero-sequence alarm information, recorded waveform information, voiceprint information and environmental information.

[0068] S2. Extract risk factors from the risk characteristic information, and determine the risk level of the risk characteristic information based on the risk factors and the risk value.

[0069] As an optional embodiment, in S2, extracting the risk factor in the risk feature information includes the following steps:

[0070] The number of alarms and alarm trends in the zero-sequence alarm information within the sampling time scale are obtained, and the alarm risk factor is determined based on the alarm trend, the number of alarms and the normalization algorithm.

[0071] Specifically, the warning risk factor ARF The formula expression form is:

[0072] ; ;

[0073] Where n is the number of time periods in which instantaneous zero-sequence alarms occur within the sampling time scale, is the changing trend of the number of alarms, and Respectively represent the number of alarms in these two time windows, is the trend weight coefficient, which is used to adjust the impact of the alarm trend in the risk factor. max( A ) is the maximum number of alarms within the sampling time scale.

[0074] This embodiment comprehensively considers the number of alarms and their changing trends. Specifically, by calculating the changing trend of the number of alarms, the increase or decrease of alarm activities can be monitored in real time, so as to timely discover potential safety hazards and more accurately reflect the alarm risk level of the cable line; if A significant increase indicates that the alarm activity increases sharply in a short period of time, which may mean that the cable insulation degradation is accelerated. The system can issue an early warning so that timely measures can be taken.

[0075] As an optional embodiment, in S2, the step of extracting risk factors from risk feature information further includes the following steps:

[0076] The number of zero-sequence spikes, the amplitude of zero-sequence spikes, the duration of zero-sequence spikes, the trend of spike mutations, and the comprehensive judgment results of the amplitude of zero-sequence spikes with the same mother at the same time in the waveform information;

[0077] The spike risk factor is determined based on the number of zero-sequence spikes, the amplitude of zero-sequence spikes, the duration of zero-sequence spikes, the amplitude of zero-sequence spikes with the same mother at the same time, and the spike mutation trend combined with the normalization algorithm;

[0078] Among them, the comprehensive analysis results of the zero-sequence spike amplitude of the same bus at the same time include: obtaining the instantaneous zero-sequence alarm set of all outgoing lines of the same bus within the sampling time scale and comparing the amplitudes of the alarm zero-sequence waveforms; determining the alarm interval corresponding to the line with the maximum amplitude of the zero-sequence waveform to increase the same-bus alarm weight.

[0079] Specifically, the spike risk factor SRF The formula is as follows:

[0080] ; ;

[0081] in: N is the number of zero-sequence spikes, and Respectively represent the number of zero-sequence spikes in the i-th and i+1-th recorded waveforms; is the maximum value of the zero-sequence spike amplitude; D is the duration of the zero-sequence spike; Δ S It is the quantitative value of the spike mutation trend. G represents the same-mother alarm weight value, which is set according to actual needs. It reflects the changing trend of parameters such as the number, amplitude or duration of zero-sequence spikes within the sampling time scale. It is of great significance for identifying potential line insulation hazards and can be calculated by comparing the changes in the amplitude or duration of adjacent spikes. and represent the amplitude and duration of the i-th spike respectively.

[0082] In this embodiment, by comprehensively analyzing the number, amplitude, duration and mutation trend of spikes, the system can more accurately assess the spike risk in the cable line, help discover potential insulation degradation phenomena, and take timely measures to prevent faults.

[0083] As an optional embodiment, in S2, the step of extracting risk factors from risk feature information further includes the following steps:

[0084] Acquire the voiceprint feature information in the voiceprint information within the sampling time scale, and the time adaptability between the voiceprint mutation moment and the instantaneous zero-sequence current alarm moment;

[0085] The voiceprint risk factor is determined based on the voiceprint feature information, the adaptability of the voiceprint mutation and the instantaneous zero-sequence current alarm moment combined with the normalization algorithm.

[0086] Specifically, voiceprint risk factors VRF The mathematical formula is as follows:

[0087] ;

[0088] Where n is the total number of voiceprint events within the sampling time scale; Represents the characteristic information of the i-th voiceprint event (comprehensive evaluation of frequency, amplitude, and duration); It represents the time adaptability between the ith voiceprint mutation and the instantaneous zero-sequence current alarm moment. and is the weight coefficient, which is used to adjust the influence of voiceprint characteristics and time adaptation in risk factors. and ; By analyzing the compatibility between the voiceprint information and the instantaneous zero-sequence current alarm moment, the system can discover the potential correlation between the voiceprint change and the cable fault; it helps to warn of cable faults in advance and improve the accuracy and efficiency of fault handling.

[0089] As an optional embodiment, the step of extracting risk factors from risk feature information further includes the following steps:

[0090] Obtain the humidity change trend in the environmental information within the sampling time scale and the proportion of instantaneous zero-sequence alarms occurring in high humidity periods;

[0091] The temperature and humidity risk factors are determined based on the humidity change trend and the proportion of instantaneous zero-sequence alarms occurring at high humidity moments combined with a normalized algorithm.

[0092] Specifically, the mathematical formula of the temperature and humidity risk factor is as follows:

[0093] ;

[0094] Where: n is the total number of sampling points within the sampling time scale; It is Humidity variation trend of each sampling point; is the proportion of instantaneous zero-sequence alarms at the i-th sampling point occurring during the high humidity period; Indicates the maximum humidity within the sampling point. The maximum value of the proportion of high humidity periods within the sampling point.

[0095] As an optional embodiment, in S2, determining the risk level of the risk feature information according to the risk factor combined with the risk value includes the following steps:

[0096] The IV value algorithm is used to obtain the risk value of each characteristic attribute information relative to historical hidden dangers;

[0097] The risk value of risk characteristic information is determined by weighted summation of risk value and risk factor;

[0098] The degree of risk is determined by comparing the risk value with the risk level range.

[0099] In this embodiment, the IV value (Information Value) algorithm is used to obtain the risk value of each characteristic attribute information relative to the historical hidden dangers. The IV value algorithm is usually used in the fields of feature selection, model construction and variable importance assessment. In this embodiment, all hidden danger events and their related characteristic attribute information that occurred in the past period of time are first collected to form a historical hidden danger data set. Determine the characteristic attributes related to the hidden danger event, such as the number of zero-sequence alarms, waveform characteristics of the recorded wave, comprehensive analysis of the waveform amplitude of the same mother line, voiceprint characteristics, environmental humidity, etc.; calculate the WOE (weight of evidence) after the characteristic attributes are binned. The IV value is obtained by weighted summing the WOE values ​​of all boxes, and the weight is the proportion of positive and negative examples in each box to the total positive and negative examples; the IV value reflects the predictive ability of the characteristic attribute for the target variable (i.e., whether a hidden danger occurs). The characteristic attributes with higher IV values ​​should be given greater weights in risk assessment because they play a more important role in identifying hidden dangers.

[0100] As an optional embodiment, the mathematical formula of the risk value R can be expressed as:

[0101] ;

[0102] in, Indicates risk factors (in this embodiment, the four risk factors calculated above are alarm risk factor, spike risk factor, voiceprint risk factor, and temperature and humidity risk factor); Indicates the risk value corresponding to the j-th risk factor (obtained through the IV value algorithm); represents the weight of the i-th risk factor; R: risk value (obtained by weighted summation); L represents the risk level (determined by comparing the risk value with the risk level interval); in step S1, we have obtained the risk value of each characteristic attribute information (i.e., risk factor) relative to the historical hidden dangers through the IV value algorithm .

[0103] Furthermore, the risk level L is determined by comparing the risk value R with the risk level interval. This comparison process is usually a classification or piecewise function. , is a function defined based on the risk level range. For example, if the risk level range is defined as low, medium, and high, then It can be a piecewise function that determines the risk level L as low, medium or high depending on the value of R. Therefore, the mathematical formula for the risk level is:

[0104] .

[0105] In this embodiment, the risk level can more accurately reflect the actual safety status of the cable line through a comprehensive assessment of multiple risk factors and combined with the IV value algorithm of historical hidden danger data; through real-time dynamic monitoring and evaluation of the risk level, the system can promptly discover and deal with potential safety hazards, prevent further deterioration and expansion of hidden dangers, avoid the limitations of single-dimensional data analysis, and improve the accuracy and reliability of hidden danger identification.

[0106] S3. Determine potential hazard points based on the risk level and their corresponding positioning strategies, and formulate corresponding potential hazard disposal measures.

[0107] like Figure 3 Specifically, the steps include:

[0108] S31. Compare and select risk characteristic information with hidden dangers based on risk levels;

[0109] S32, determining a potential danger point according to source attribute information corresponding to risk characteristic information of a potential danger; determining a danger level corresponding to the potential danger point according to a network topology;

[0110] S33, determining the maintenance priority according to the risk level and the corresponding critical situation level to obtain a corresponding maintenance strategy;

[0111] S34. The execution carrier performs maintenance actions in response to the maintenance strategy.

[0112] It can be understood that in step S2, by integrating multiple risk factors (such as alarm risk factor, spike risk factor, voiceprint risk factor, temperature and humidity risk factor, etc.) and risk value, the risk value of the risk characteristic information of each bus or branch is determined, and it is compared with the preset risk level interval to determine the risk level. At the beginning of step S3, the system will compare these risk levels and screen out risk characteristic information with hidden dangers. Specifically, the system will mark the risk characteristic information with a risk level higher than the preset threshold as having hidden dangers; for the risk characteristic information with hidden dangers screened out, the system will determine the specific location of the hidden danger point according to its corresponding source attribute information; the source attribute information usually includes key information such as the identification and location of the bus or branch, through which the system can accurately locate the location of the hidden danger point in the network topology map; assuming that in step S2, the system determines that a branch named "L1" has a high risk hidden danger by analyzing the zero-sequence alarm information, recording waveform information, voiceprint information and environmental information. In step S3, the system first parses the source attribute information of this hidden danger information and finds that it belongs to the outlet branch of station A. Then, the system locates the location of station A and its export branch in the network topology map, and confirms that the "L1" branch is the potential danger point. Finally, the system marks the "L1" branch as a potential danger point in the network topology map, and records the relevant risk level and maintenance suggestions; after determining the potential danger point and its degree of danger, the system will determine the maintenance priority based on the risk level and the degree of danger. Specifically, the higher the risk level and the more serious the degree of danger of the potential danger point, the higher its maintenance priority. The system will formulate corresponding maintenance strategies based on the maintenance priority, including arrangements for maintenance time, maintenance personnel, maintenance tools, etc. The execution carrier (such as inspection robots, maintenance personnel, etc.) will respond to the maintenance strategy formulated by the system and perform corresponding maintenance actions. During the execution process, the system will also monitor the maintenance progress and effect in real time, and adjust and optimize the maintenance strategy according to the actual situation.

[0113] It is understandable that for a high-priority potential risk point, the system may dispatch an inspection robot to the site for a preliminary inspection and formulate a detailed maintenance plan based on the inspection results; at the same time, the system will evaluate the maintenance effect through real-time monitoring to ensure that the potential risk is completely eliminated. Through the above steps, the system can accurately determine the potential risk point and formulate scientific and reasonable potential risk disposal measures according to the risk level and its corresponding positioning strategy. This method not only improves the accuracy and efficiency of fault handling, but also reduces the risk of further expansion of safety hazards, providing a strong guarantee for the safe and stable operation of the power grid.

[0114] Embodiment 2: Since the calculation of the spike risk factor is related to multiple factors, by comprehensively analyzing the number, amplitude, duration and mutation trend of the spikes, the system can more accurately assess the spike risk in the cable line, which helps to discover potential insulation degradation phenomena. Another way to calculate the spike risk factor is as follows: SRF = the average value of the maximum amplitude of the zero-sequence spikes in the most recent N recording files Weight + average duration of zero-sequence spike Weight The average number of zero-sequence spikes in the most recent N recording files The weight + the changing trend of the above three dimensions (this period - the last period / max (this period - the last period)) + the waveform with the largest amplitude in the comprehensive judgment of the zero-sequence spike wave of the same mother at the same time increases the alarm weight of multiple lines of the same mother; for example, the above N can be 5, and the above weights are adjusted according to actual needs and are not limited here.

[0115] Embodiment 3: A technical solution also provided in the embodiment of the present invention is a line insulation hidden danger monitoring system, such as Figure 4 As shown, including:

[0116] Data acquisition unit 101: acquires risk characteristic information of each busbar and / or branch line in the distribution network automation network in real time according to the risk stress mechanism;

[0117] Analysis unit 102: extract risk factors from risk feature information, and determine the risk level of the risk feature information based on the risk factors and risk value;

[0118] Execution unit 103: Determine the potential danger point according to the risk level and its corresponding positioning strategy, and formulate corresponding potential danger disposal measures.

[0119] The line insulation hidden danger monitoring system collects and analyzes data through three data processing methods. The following is an explanatory explanation through specific working conditions or scenarios.

[0120] (1) All signals collected by the information collectors installed on the busbar or branch line are sent to the main station for processing. The main station analyzes the four sub-dimensional data (specifically: zero-sequence alarm information, recorded waveform information, voiceprint information and environmental information), calculates the risk value of each signal, and locates the fault position through the end of the continuous signal greater than the risk value threshold. The 10kV outgoing line switch of the main grid 110kV substation can also be configured in the same way to capture the cable discharge between the substation and the first 10kV disconnection point.

[0121] (2) Perform local calculations through enhanced DTUs, increase the existing zero-sequence sampling frequency through zero-sequence loop Hall sensors and high-frequency sampling to more accurately analyze zero-sequence waveforms, and simultaneously add voiceprint sensors and humidity sensors. Further judgments are made through DTU's recording analysis of voiceprint telemetry and fitting of zero-sequence overcurrent alarm information. DTUs must be configured with 0s delay and 10-40A zero-sequence overcurrent alarms. It is recommended that cable lines below 10km use this setting. Longer lines can also use 3A per kilometer. Sensitivity is adjusted. Another way is to extract more zero-sequence alarms through 0s delay and 5A zero-sequence overcurrent alarm, and make detailed judgments with the amplitude corresponding to the alarm. After DTU processing, the discharge risk value of the interval (ring in, ring out and branch line interval) of the local zero-sequence monitoring is sent to the master station through the point table, and further comparison and positioning are performed through the master station. The DTU can judge through the alarm information of the ring in and ring out that if there is a zero-sequence overcurrent alarm on the ring in line, and there is no zero-sequence overcurrent alarm on the ring out or the risk value is low, then there is a discharge risk in this station; if there are zero-sequence overcurrent alarms on both the ring in and ring out lines, and the risk value is high, then there is a corona discharge risk in the station or line on the load side of this station.

[0122] (3) Perform local calculations through the original DTU peripheral module with edge computing function. The module obtains DTU telemetry data through 485 / 232 or serial communication. The synchronization module adds voiceprint sensors and humidity sensors. After the module calculates the risk value of the four-dimensional data, it returns the result to the DTU. The DTU sends the discharge risk value of the interval (ring in, ring out or branch line interval) of the local zero-sequence monitoring through the point table and sends it to the main station for further comparison and positioning. The DTU can judge from the alarm information of the ring in and ring out that if there is a zero-sequence overcurrent alarm on the ring in line, but there is no zero-sequence overcurrent alarm on the ring out or the risk value is low, then there is a risk of corona discharge at this station; if there is a zero-sequence overcurrent alarm on both the ring in and ring out, and the risk value is high, then there is a risk of corona discharge at the station or line on the load side.

[0123] The line insulation hidden danger monitoring system also includes two on-site verification methods used when hidden danger events occur; the following is an explanatory explanation through specific working conditions or scenarios.

[0124] (1) After positioning, the master station will send the positioning information exceeding the risk value to the device owner for partial discharge live detection. After the interval positioning exceeds 20db, power outage and maintenance will be carried out.

[0125] (2) The master station sends the located interval (a section of bus where the DTU is located, or a branch interval with zero-sequence monitoring) to the DTU (remote control, remote signaling) through the point table, activates the switch station partial discharge monitoring robot through the remote control loop, and communicates with the robot through the DTU (micro-power wireless or lora) to inform the robot of the interval or bus that needs to be inspected, and returns the partial discharge value and the corresponding interval information to the DTU as the inspection result, which is sent to the master station through the DTU. This method can avoid the inaccurate data caused by the time difference of partial discharge detection (the discharge time is closely related to the operating environment at the time, and manual detection often misses the best detection time). Intervals greater than 20DB or where the partial discharge value increases significantly will be subject to power outage maintenance.

[0126] In this embodiment, first, the risk characteristic information of each busbar or / and branch line in the distribution network automation network is obtained in real time according to the risk stress mechanism. By collecting these multi-dimensional data, a comprehensive and accurate information basis is provided for the subsequent hidden danger identification; after obtaining the risk characteristic information, the present invention extracts the risk factors therein, and determines the risk level of the risk characteristic information according to the risk factors combined with the risk value. Through the comprehensive analysis and calculation of the risk factors, the degree of safety hazards of the cable line can be more accurately evaluated. By comparing the risk levels, the risk characteristic information with hidden dangers is screened out, and the specific location of the hidden danger point is determined according to the source attribute information. At the same time, the degree of danger and degradation corresponding to the risk point is determined according to the network topology, such as the scope of hidden dangers, maintenance time, waveform mutation trend and alarm trend, so as to formulate a scientific and reasonable maintenance priority and corresponding maintenance strategy, improve the accuracy and efficiency of fault handling, and further reduce the further expansion of safety hazards.

[0127] Embodiment 4: An optional embodiment also provided in the embodiments of the present invention is: an electronic device, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the processor calls the computer program in the memory, the steps of the line insulation hidden danger monitoring method based on multi-parameter analysis are implemented.

[0128] Embodiment 5: An optional embodiment also provided in the embodiments of the present invention is: a storage medium, in which computer executable instructions are stored. When the computer executable instructions are loaded and executed by a processor, the steps of a line insulation hazard monitoring method based on multi-parameter analysis are implemented.

[0129] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be assigned to different functional modules as needed, that is, the internal structure of the specific device can be divided into different functional modules to complete all or part of the functions described above.

[0130] In the embodiments provided in the present application, it should be understood that the disclosed structures and methods can be implemented in other ways. For example, the embodiments of the structure described above are only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another structure, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, structures or units, which can be electrical, mechanical or other forms.

[0131] The units described as separate components may or may not be physically separated, and the components shown as units may be one physical unit or multiple physical units, that is, they may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0132] In addition, each functional unit in the embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0133] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium, including several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to perform all or part of the steps of the various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0134] The specific implementation described above is a preferred implementation of a line insulation hazard monitoring method and system based on multi-parameter analysis of the present invention, and is not intended to limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation. All equivalent changes made in accordance with the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A line insulation hidden danger monitoring method based on multi-parameter analysis, characterized in that: The steps include: S1. Obtain risk characteristic information of each busbar and / or branch line in the distribution network automation monitoring network in real time according to the risk stress mechanism; S2. extracting risk factors from the risk characteristic information, and determining the risk level of the risk characteristic information based on the risk factors and the risk value; S3. Determine the potential danger points according to the risk level and their corresponding positioning strategies, and formulate corresponding potential danger disposal measures; In S2, the step of extracting risk factors from risk feature information further includes the following steps: Obtain the number of zero-sequence spikes, zero-sequence spike amplitude, duration of zero-sequence spikes, spike mutation trend, and comprehensive judgment results of zero-sequence spike amplitudes of the same mother at the same time in the recorded waveform information corresponding to the zero-sequence alarm that returns instantaneously within the sampling time scale; The spike risk factor is determined based on the number of zero-sequence spikes, the amplitude of zero-sequence spikes, the duration of zero-sequence spikes, the amplitude of zero-sequence spikes with the same mother at the same time, and the spike mutation trend combined with the normalization algorithm; Among them, the comprehensive analysis results of the zero-sequence spike amplitude of the same bus at the same moment include: obtaining the instantaneous zero-sequence alarm set of all outgoing lines of the same bus within the sampling time scale and comparing the amplitudes of the alarm zero-sequence waveforms; determining the alarm interval corresponding to the line with the maximum amplitude of the zero-sequence waveform to increase the same-bus alarm weight; Among them, the spike risk factor The formula is as follows: ； in: N is the number of zero-sequence spikes, N i and N i+1 Respectively expressed in i and i +The number of zero-sequence spikes in 1 recorded waveform; W avg is the maximum value of the zero-sequence spike amplitude; D is the duration of the zero-sequence spike; Δ S is the quantitative value of the sharp wave mutation trend, G Indicates the weight value of the same mother alarm, which can be set according to actual needs; and Respectively represent i The amplitude and duration of the spike.

2. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 1 is characterized in that: In S1, risk characteristic information of each busbar and / or branch line in the distribution network automation monitoring network is obtained in real time according to the risk stress mechanism; The steps include: A risk stress mechanism is set according to the waveform characteristics of the zero-sequence current. When the risk stress mechanism is responded to, risk characteristic information of each busbar and / or branch line of the distribution network automation is obtained; The risk characteristic information includes source attribute information, time attribute information and characteristic attribute information.

3. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 2 is characterized in that: The characteristic attribute information at least includes instantaneous reset zero-sequence alarm information, recorded waveform information, soundprint information and environmental information.

4. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 1 or 3, characterized in that: In S2, the step of extracting risk factors from risk feature information includes the following steps: Obtain the alarm times and alarm trends in the zero-sequence alarm information of instantaneous return within the sampling time scale; The alarm risk factor is determined based on the alarm trend, the number of alarms and the normalization algorithm.

5. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 1 or 3, characterized in that: In S2, the step of extracting risk factors from risk feature information further includes the following steps: Acquire the voiceprint feature information in the voiceprint information within the sampling time scale, and the time adaptability between the voiceprint mutation moment and the instantaneous zero-sequence current alarm moment; The voiceprint risk factor is determined based on the voiceprint feature information, the adaptability of the voiceprint mutation and the instantaneous zero-sequence current alarm moment combined with the normalization algorithm.

6. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 1 or 3 is characterized in that: In S2, the step of extracting risk factors from risk feature information further includes the following steps: Obtain the humidity change trend in the environmental information within the sampling time scale and the proportion of zero-sequence alarms with instantaneous return occurring in high humidity periods; The temperature and humidity risk factors are determined based on the humidity change trend and the proportion of instantaneous zero-sequence alarms occurring at high humidity moments combined with a normalized algorithm.

7. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 2 is characterized in that: In S2, the step of determining the risk level of the risk characteristic information based on the risk factor combined with the risk value includes the following steps: The IV value algorithm is used to obtain the risk value of each characteristic attribute information relative to historical hidden dangers; The risk value of risk characteristic information is determined by weighted summation of risk value and risk factor; The degree of risk is determined by comparing the risk value with the risk level range.

8. The line insulation hidden danger monitoring method based on multi-parameter analysis according to claim 7 is characterized in that: In S3, the step of determining the potential danger points according to the risk level and the corresponding positioning strategies, and formulating corresponding potential danger disposal measures, includes the following steps: S31. Compare and select risk characteristic information with hidden dangers based on risk levels; S32, determining the potential danger point according to the source attribute information corresponding to the risk characteristic information of the potential danger; Determine the degree of danger corresponding to the potential risk point based on the network topology; S33, determining the maintenance priority according to the risk level and the corresponding critical situation level to obtain a corresponding maintenance strategy; S34. The execution carrier performs maintenance actions in response to the maintenance strategy.

9. A line insulation hidden danger monitoring system, applicable to the line insulation hidden danger monitoring method based on multi-parameter analysis according to any one of claims 1 to 8, characterized in that: Data acquisition unit: acquires risk characteristic information of each bus or branch line in the distribution network automation monitoring network in real time according to the risk stress mechanism; Analysis unit: extract risk factors from risk feature information, and determine the risk level of risk feature information based on the risk factors and risk value; Execution unit: Determine potential danger points based on the risk level and their corresponding positioning strategies, and formulate corresponding potential danger disposal measures.

10. An electronic device, characterized in that: It comprises a memory and a processor, wherein the memory stores a computer program, and when the processor calls the computer program in the memory, the steps of the line insulation hidden danger monitoring method based on multi-parameter analysis as described in any one of claims 1 to 8 are implemented.

11. A storage medium, characterized in that: The storage medium stores computer executable instructions, and when the computer executable instructions are loaded and executed by the processor, the steps of the line insulation hidden danger monitoring method based on multi-parameter analysis as described in any one of claims 1 to 8 are implemented.

Images