A Distribution Network Fault Transfer Early Warning Method and System Based on Multidimensional Fault Transfer Risk Indicators

Abstract

This invention discloses a distribution network fault transfer early warning method and system based on multi-dimensional fault transfer risk indicators. The method includes: acquiring and fusing multi-dimensional fault transfer risk indicator data of the distribution network, such as electrical quantities, equipment status, operating environment, operation and maintenance management, and user side, to form a panoramic status information; based on the panoramic status information, calculating a dynamic fault transfer risk index through a multi-level fusion assessment model that includes basic risk aggregation, coupling effect correction, and network structure vulnerability assessment; and triggering graded early warnings and generating corresponding disposal decision suggestions based on the value of the fault transfer risk index. This invention achieves panoramic perception of distribution network vulnerabilities by integrating multi-dimensional risk indicators, improves the accuracy of risk quantification through dynamic fusion assessment, and achieves advanced early warning of risk evolution patterns by matching a fault precursor feature library, ultimately forming an operation and maintenance closed loop from risk perception to proactive disposal, significantly enhancing the proactive defense capability of the distribution network.

Description

Technical Field

[0001] This invention relates to the field of power distribution network fault early warning and safety protection technology, specifically to a power distribution network transfer early warning method and system based on multi-dimensional fault transfer risk indicators. Background Technology

[0002] As a crucial link connecting the main grid and users, the reliability of the distribution network directly affects power supply security. In recent years, with the expansion of the power grid and the increase in equipment complexity, "multi-damage, multi-explosion" type faults have occurred frequently. The core characteristic of these faults is "fault transfer," meaning that an initial fault point is not isolated in a timely and effective manner, leading to a change in the fault current path and subsequently triggering cascading faults in adjacent lines or equipment, greatly expanding the scope of the accident. Analysis shows that fault transfer is not caused by a single reason, but is the result of the combined effects of multiple fault transfer risk indicators, including automation terminals, cable bodies, operating environment, and user side. Specifically, automation terminals suffer from incomplete functions, incorrect settings, or low coverage, making it impossible to quickly and accurately locate and isolate the initial fault; cable systems have weak insulation points due to manufacturing defects, accessory installation quality problems, and insufficient operation and maintenance; in addition, harsh operating environments and the failure to promptly rectify hidden dangers in user equipment further exacerbate the system's operational risks. These transfer risk indicators are coupled with each other, allowing a simple single-phase ground fault to rapidly develop into a phase-to-phase short circuit or a three-phase short circuit, ultimately exceeding the protection range and causing grid paralysis.

[0003] Currently, fault early warning technologies in distribution networks largely rely on monitoring the exceedances of electrical quantities such as voltage and current, which is a relatively delayed and passive response. Existing methods lack a systematic and forward-looking comprehensive detection and assessment of the aforementioned multidimensional and potential non-electrical transfer risk indicators. Therefore, they cannot identify high-risk states of the system in the early stages of fault chain evolution, making it difficult to effectively prevent fault transfer. Thus, there is an urgent need for a new technology that can deeply integrate multidimensional fault transfer risk indicator information to achieve early and accurate early warning of distribution network fault transfer risks, thereby transforming passive handling into proactive defense and fundamentally improving the resilience and safe operation level of the distribution network. Summary of the Invention

[0004] This invention provides a distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators, which solves the shortcomings of traditional methods that rely on electrical quantities, fail to adequately consider multi-dimensional potential risk indicators such as equipment status and operating environment, and suffer from delayed risk assessment. This invention is achieved through the following technical solution:

[0005] Firstly, this application provides a distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators, comprising the following steps:

[0006] Acquire and integrate multi-dimensional fault transfer risk indicator data of the distribution network to form a panoramic status information;

[0007] Based on the panoramic status information, a dynamic failover risk index is calculated using a multi-level fusion evaluation model.

[0008] Based on the value of the failover risk index, a tiered early warning is triggered and corresponding handling decision recommendations are generated.

[0009] A further optimized solution involves acquiring and integrating multi-dimensional fault transfer risk indicator data of the distribution network to form panoramic status information, specifically including the following steps:

[0010] Collect real-time electrical quantity data, equipment status data, and operating environment data of the power distribution network;

[0011] Obtain operation and maintenance management data of the power distribution network and user-side status data;

[0012] By performing time alignment and spatial correlation on the above multi-source heterogeneous data, a panoramic status information of multi-dimensional fault transfer risk indicators under a unified spatiotemporal benchmark is constructed.

[0013] A further optimized solution is that the equipment status data collected from the real-time electrical quantity data, equipment status data, and operating environment data of the power distribution network includes:

[0014] The functional operation status and setting information of the power distribution automation terminal are obtained;

[0015] Sensors deployed at cable joints are used to monitor partial discharge signals and joint temperature online.

[0016] A further optimization scheme involves calculating a dynamic failover risk index based on the panoramic state information using a multi-level fusion evaluation model, specifically including the following steps:

[0017] In the first basic risk aggregation layer, the standardized risk values ​​of each independent failover risk indicator are calculated and aggregated to obtain the basic risk value;

[0018] In the second coupling effect correction layer, the coupling correction coefficients for the interaction between different categories of failover risk indicators are calculated;

[0019] In the third network structure vulnerability assessment layer, the network structure risk coefficient is calculated based on real-time power grid topology and power flow.

[0020] By combining the basic risk value, coupling correction coefficient, and network structure risk coefficient, a dynamic failover risk index is calculated.

[0021] A further optimization scheme is proposed, namely, the failover risk index. The calculation formula is shown below:

[0022] ;

[0023] in, For the first The dynamic weight of each basic risk. Its basic risk value, In order to be with the first The product of all coupling correction coefficients related to the basic risk, This represents the network structure risk coefficient.

[0024] A further optimized approach is that the calculation of the standardized risk value includes:

[0025] For continuously monitored time-domain and frequency-domain indicators, a hybrid normalization method based on threshold and statistical distribution is used for conversion;

[0026] For discontinuous event-based status indicators, convert them into risk values ​​with a time-decrease factor.

[0027] A further optimization involves introducing a dynamic weight adjustment mechanism during the calculation of the dynamic failover risk index. The improvement is based on the matching degree between real-time indicator data and a pre-built fault precursor feature library, and the calculation formula is as follows:

[0028] ;

[0029] in, For static base weights, As an enhancing factor, For real-time matching accuracy.

[0030] A further optimized solution is that the tiered early warning system is a three-level early warning system, including:

[0031] When the failover risk index is greater than or equal to the first preset threshold and less than the second preset threshold, a first-level warning is triggered and an inspection suggestion is generated.

[0032] When the failover risk index is greater than or equal to the second preset threshold and less than the third preset threshold, a second-level warning is triggered, and a warning report containing the dominant risk indicators and targeted handling suggestions is generated.

[0033] When the failover risk index is greater than or equal to the third preset threshold, a third-level warning is triggered, and alarm information and a pre-decision isolation scheme are generated.

[0034] The first preset threshold, the second preset threshold, and the third preset threshold increase sequentially.

[0035] The further optimized plan is that the handling recommendations in the second-level early warning report include conducting special testing or verifying the protection settings within a specified time.

[0036] Secondly, this application provides a distribution network fault transfer early warning system based on multi-dimensional fault transfer risk indicators, including:

[0037] The multi-dimensional fault transfer risk indicator perception and data fusion module is used to acquire multi-source heterogeneous data from power distribution automation terminals, sensors deployed in cables and the environment, production management systems and user management systems, and perform time alignment and spatial correlation processing on the data to output panoramic status information of multi-dimensional fault transfer risk indicators.

[0038] The multi-level dynamic fusion risk assessment module is communicatively connected to the multi-dimensional failover risk index perception and data fusion module. It is used to receive the panoramic state information, perform fusion calculations through basic risk aggregation, coupling effect correction and network structure vulnerability assessment, and output a dynamic failover risk index.

[0039] The graded early warning and response decision support module is communicatively connected to the multi-level dynamic fusion risk assessment module. It is used to receive the fault transfer risk index, classify it according to a preset threshold, and generate response decision suggestions corresponding to the early warning level.

[0040] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0041] By integrating multi-dimensional electrical and non-electrical risk indicators to comprehensively perceive the vulnerabilities of the distribution network, a multi-level dynamic fusion assessment model is constructed to accurately quantify fault transfer risks. This model matches real-time data with a fault precursor feature database to achieve proactive early warning of risk evolution patterns. Ultimately, through the linkage of tiered early warning and targeted handling recommendations, a closed-loop operation and maintenance system is formed, from risk perception to proactive handling. (See attached diagram.)

[0042] To more clearly illustrate the technical solutions of the exemplary embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of the present invention and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0043] Figure 1 A flowchart of a distribution network transfer early warning method based on multi-dimensional fault transfer risk indicators provided in this application embodiment;

[0044] Figure 2 Another flowchart of the distribution network transfer early warning method based on multi-dimensional fault transfer risk indicators provided in the embodiments of this application;

[0045] Figure 3 This is a functional block diagram of a distribution network transfer early warning system based on multi-dimensional fault transfer risk indicators, provided in an embodiment of this application. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the embodiments and accompanying drawings. The illustrative embodiments and descriptions of the present invention are only used to explain the present invention and are not intended to limit the present invention.

[0047] First, some of the technical terms used in this application will be explained to help those skilled in the art understand this application.

[0048] DTU: Distribution Terminal Unit;

[0049] FTU: Feeder Terminal Unit;

[0050] PD: Partial Discharge;

[0051] UHF: Ultra-High Frequency, extra high frequency;

[0052] HFCT: High-Frequency Current Transformer;

[0053] PMS: Production Management System;

[0054] PRPD: Phase-Resolved Partial Discharge;

[0055] THD: Total Harmonic Distortion;

[0056] GIS: Geographic Information System;

[0057] FTR: Fault Transfer Risk;

[0058] R: Risk;

[0059] K: Coupling coefficient;

[0060] S: Structural vulnerability coefficient.

[0061] This invention provides a distribution network fault transfer early warning method and system based on multi-dimensional fault transfer risk indicators. Addressing the shortcomings of traditional early warning methods that rely on electrical quantities and neglect potential fault transfer risk indicators such as equipment status and environment, this invention constructs a perception system encompassing multi-dimensional fault transfer risk indicators including automation terminals, cable joints, operating environment, and user side. A simulation feature library is established to simulate the chain-like evolution of faults, and a multi-level dynamic fusion risk assessment model is proposed to quantify the coupling effect between fault transfer risk indicators and network operating status, enabling proactive risk assessment. Finally, through a tiered early warning mechanism, an actionable report containing dominant fault transfer risk indicators, transfer paths, and handling suggestions is output, forming a closed loop from risk perception to proactive handling, thereby improving the distribution network's ability to prevent fault transfer.

[0062] Firstly, such as Figure 1 and Figure 2 As shown, this application provides a distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators, including the following steps:

[0063] Step S1: Acquire and integrate multi-dimensional fault transfer risk indicator data of the distribution network to form panoramic status information;

[0064] Step S2: Based on the panoramic state information, a dynamic fault transfer risk index is calculated through a multi-level fusion evaluation model; wherein, in the process of calculating the fault transfer risk index, the weight of each risk index is dynamically adjusted according to the matching degree between real-time indicators and the pre-constructed fault precursor feature library, so as to achieve sensitive identification and early warning of specific risk evolution patterns.

[0065] Step S3: Based on the value of the failover risk index, trigger a graded early warning and generate corresponding handling decision suggestions.

[0066] This embodiment integrates heterogeneous data from multiple sources, including electrical quantities, equipment status, environment, operation and maintenance, and users, and performs spatiotemporal alignment and fusion to construct a panoramic status information covering "equipment-environment-operation and maintenance." This enables comprehensive perception of the distribution network's risk status. Furthermore, utilizing a multi-level fusion assessment model, it comprehensively aggregates basic risks, corrects coupling effects, and assesses network structural vulnerability to dynamically calculate an accurate fault transfer risk index, improving the accuracy and real-time performance of risk assessment. Finally, based on this index, it triggers tiered early warnings and automatically generates differentiated handling decision recommendations, forming a closed-loop management system from proactive inspection to emergency isolation. This effectively enhances the distribution network's early warning capability for potential fault transfers and its operation and maintenance response efficiency, ensuring the safe and stable operation of the power grid.

[0067] In one embodiment, step S1: acquiring and fusing multi-dimensional fault transfer risk index data of the distribution network to form panoramic status information specifically includes the following steps:

[0068] Step S11: Collect real-time electrical quantity data of the power distribution network;

[0069] Specifically, voltage transformers and current transformers are installed at the 10kV outgoing switchgear, ring main unit incoming and outgoing line units, cable branch boxes, and high-voltage side outlets of distribution transformers in the substation to measure the voltage of each phase (U). a U b U c ) and current (I) a , I b ,I c The instantaneous waveform and RMS value of the signal are used to calculate power, monitor load, and identify phase-to-phase short circuits.

[0070] At the beginning of the cable line (such as the substation outgoing line) and multiple segment points, the zero-sequence current (3I0) is measured by a zero-sequence current transformer, and the zero-sequence voltage (3U0) is obtained by the open delta winding of the voltage transformer or a special sensor.

[0071] By sampling the current waveform at high speed, the rate of change of current (di / dt) and the over-limit amplitude at the moment the fault occurs are calculated, which are used to determine the nature of the fault (such as metallic grounding or arc grounding) and assess the severity of the fault.

[0072] Step S12: Collect equipment status data and operating environment data of the power distribution network;

[0073] Specifically, the condition of the cable itself and its accessories is monitored:

[0074] Ultra-high frequency (UHF) sensors or high-frequency current transformers (HFCTs) are installed on the outer wall of weak insulation points such as intermediate joints and terminations of cables to monitor the amplitude, pulse frequency, and phase distribution (PRPD spectrum) of partial discharge (PD) signals online. This indicator directly reflects the activity level of internal insulation defects (such as air gaps and impurities) or joint installation process defects (such as stress cone misalignment and improper treatment of semiconductor layer fractures).

[0075] Install a wireless temperature sensor or lay a distributed fiber optic temperature measurement system close to the cable joint to continuously monitor the operating temperature and temperature rise rate of the joint surface. Abnormal temperature rise is a direct manifestation of excessive contact resistance, reduced current carrying capacity, or internal insulation deterioration causing heat generation.

[0076] For directly buried or tunnel cables, distributed fiber optic acoustic or strain sensing systems are installed in sections of the cable body that are susceptible to external damage (such as near construction areas) and at joints to monitor the cable's vibration, deformation, or stress changes, in order to provide early warning of external damage risks and assess the mechanical stability of joints.

[0077] Monitoring the status of distribution automation terminals: The distribution master station system periodically queries or receives information reported by the terminals, such as the activation / deactivation status of the function soft switch, the current protection setting group number and value, the signal strength of the communication status, and the equipment self-test abnormal codes, in order to quantitatively evaluate the functional integrity, setting accuracy, and online reliability of the terminals;

[0078] Monitor operating environment parameters:

[0079] Temperature and humidity sensors are installed inside cable trenches, tunnels, pipe racks, and ring main unit foundation compartments to measure ambient temperature and relative humidity. High humidity is a major risk factor for accelerating cable insulation aging and causing water treeing.

[0080] Install water level sensors in low-lying areas of enclosed cable channels;

[0081] In areas with fire risk (such as where multiple cables are laid in the same trench), install combustible gas (such as methane) or smoke concentration sensors to provide early warning of severe environmental conditions that may cause fire or explosion.

[0082] Step S13: Integrate user load, hidden danger rectification records, equipment test records, and other operation and maintenance and user-side management data from electricity information collection, production management (PMS), and marketing systems;

[0083] Specifically, the system obtains information on three-phase load imbalance and overcapacity operation alarms from the user's electricity consumption information collection system; it accesses various operation and maintenance management data for cables and accessories from the production management system (PMS), including the most recent preventive test records (such as the results of withstand voltage tests and partial discharge tests), construction acceptance reports, and construction permit information for external damage protection zones; and it obtains feedback status of notifications regarding user hazard rectification and their final completion status from the marketing system.

[0084] Step S14: The multi-source heterogeneous data collected and integrated in steps S11, S12 and S13 are fused to form panoramic status information;

[0085] Specifically, by assigning a unified and accurate timestamp to all sensor data and status information, and associating it with the spatial coordinates of equipment in the power grid resource management system (such as the locations of poles, stations, and trenches), unified benchmark multi-source data that is accurately aligned in both time and space dimensions is obtained.

[0086] Secondly, based on the extended public information model, a digital object is created for each physical device (such as a cable, a connector, or a terminal), that is, its digital twin is constructed, and each digital twin is associated and bound with its corresponding real-time measurement data, historical status records, and operation and maintenance files to form a complete device information view;

[0087] Finally, feature extraction and standardization are performed on the original measurement data, and the normalized results are unified indicators that can be directly used by the upper-level risk assessment model.

[0088] In one embodiment, step S14, which involves feature extraction and standardization of the raw measurement data, specifically includes the following steps:

[0089] Three methods—time-domain feature extraction, frequency-domain feature extraction, and event feature extraction—are used to quantify equipment status and risk from different dimensions.

[0090] A. Time-domain feature extraction mainly includes: for continuously monitored numerical sequences x(t) (such as temperature, partial discharge intensity, voltage, current, etc.), calculating their mean, trend slope (such as linear regression slope or rate of change), and fluctuation intensity (such as standard deviation or coefficient of variation) within a sliding time window T, to characterize the steady-state level, trend, and fluctuation characteristics of the signal. The main features include the following:

[0091] Time-domain characteristics of electrical quantities: instantaneous / effective values ​​of three-phase voltage and current (U a U b U c , I a , I b , I c ); Effective / instantaneous values ​​of zero-sequence voltage and zero-sequence current (3U0, 3I0); Fault current change rate (di / dt); Negative sequence components of voltage / current; Active power, reactive power (P, Q);

[0092] Equipment status time-domain characteristics: cable joint / body temperature (T) and its temperature rise rate (dT / dt); partial discharge (PD) activity intensity, such as discharge pulse count rate and average discharge amplitude;

[0093] Operating environment time-domain characteristics: ambient temperature and relative humidity inside the cable trench / ring mains unit;

[0094] B. Frequency domain feature extraction mainly includes:

[0095] Partial discharge characteristics: For the partial discharge PRPD spectrum, calculate its discharge phase distribution entropy. This is used to assess the degree of discharge phase dispersion. An increase in entropy may indicate a change in defect type or an increased risk. The calculation formula is:

[0096] Equation (1)

[0097] in, It is the probability that the discharge pulse occurs within the power frequency phase window k.

[0098] Fault transient signal characteristics: Extract the total harmonic distortion (THD) of the current or voltage signal.

[0099] C. Event feature extraction primarily targets discontinuous state changes, operation logs, and alarm information, transforming them into time-sensitive quantitative risk indicators. Specifically, this includes the following three types of events:

[0100] Equipment function and status events: "Enable / Disable" status of automation terminal protection function; "Communication Interruption / Resumption" event of automation terminal; "Start / Action / Reset" event of protection device; "Setting Inconsistency" event; "Open Circuit / Failure" alarm event of battery cell.

[0101] Operation and maintenance and testing events: cable / joint "preventive test results" events; construction acceptance "pass / fail" events; hidden danger rectification "notification / completion / overdue" events; protected area "external damage construction permit" events.

[0102] Environmental and Abnormal Events: Cable trench "water accumulation alarm" event; fire alarm system "smoke / combustible gas" alarm event; user-side "overcapacity operation alarm" and "three-phase imbalance over-limit alarm" events.

[0103] After completing the above feature extraction, all features are normalized into a uniform dimension status index (such as health index or risk index) within the range of 0-1.

[0104] To clarify, the types of the above multidimensional fault indicators and their corresponding characteristics are summarized in Table 1:

[0105] Table 1. Types of Multidimensional Fault Indicators

[0106]

[0107] In one embodiment, step S2: calculating a dynamic failover risk index based on the panoramic state information using a multi-level fusion evaluation model, specifically includes the following steps:

[0108] Step S21: Preprocess the input original multidimensional indicators, converting them all into a unified dimensionless risk contribution, denoted as r, r∈[0,1]; for time-domain and frequency-domain indicators (such as temperature and partial discharge intensity), a hybrid normalization method based on threshold and statistical distribution is used for conversion, and its calculation formula is as follows:

[0109] Equation (2)

[0110] In the formula, The real-time value of index i (such as temperature, partial discharge pulse count, voltage imbalance). , These are safety thresholds and critical thresholds based on equipment technical standards; This is the empirical cumulative distribution function obtained based on historical normal data. To adjust the parameter and balance the absolute standard and relative statistical anomalies, a value of 0.5 was used; Results Indicates no risk. This indicates that the critical risk has been reached.

[0111] For event-based status indicators (such as "protection function deactivated" or "test failed"), they are converted into Boolean risk values ​​with a time decay factor. The calculation formula is as follows:

[0112] Equation (3)

[0113] In the formula, Let the risk value of the j-th event type be at time t. The value is 1 if the event is active (e.g., protection has not yet been restored), and 0 otherwise. The time when the event occurred or was most recently confirmed; The risk decay time constant for the j-th event type (the time constant corresponding to the "test failure" event). The time constant for a "momentary communication interruption" event is relatively long. (shorter)

[0114] Step S22: The standardized risk contribution is calculated through three layers: basic risk aggregation, coupling effect correction, and network structure vulnerability assessment. The outputs are the basic risk value, coupling correction coefficient, and network structure risk coefficient for the final comprehensive calculation. Specifically, the model includes the following three assessment layers:

[0115] In the first basic risk aggregation layer, the static basic risk formed by the superposition of various independent transfer risk indicators is calculated; specifically, similar indicators are first sub-aggregated to obtain the basic risk value. The calculation formula is:

[0116] Equation (4)

[0117] in, For the fixed base weight of the indicator, Its risk contribution;

[0118] Then, the basic risk values ​​calculated for each category are summed across categories to obtain the system's basic risk vector. ;

[0119] In the second coupling effect correction layer, the interaction (coupling, amplification effect) between different categories of risks is quantified. For this purpose, it is necessary to define and calculate the coupling correction coefficient matrix. ;

[0120] Equation (5)

[0121] In the formula, This represents the environmental humidity risk value. For the inherent risks of the equipment itself, The coupling strength coefficient (calibrated using historical data or a physical model);

[0122] Other key coupling relationships include the masking effect of terminal function anomalies on electrical quantity anomalies, which is a negative coupling, and the clustering effect caused by the simultaneous presence of defects in multiple adjacent joints, which is a positive coupling. The final output of this layer is a series of correction coefficients used for subsequent calculations;

[0123] In the third network structure vulnerability assessment layer, the network structure risk coefficient is calculated based on real-time topology and power flow data. The following formula is used to quantitatively assess the vulnerability of fault transfer under the current power grid operation mode:

[0124] Equation (6)

[0125] Where Z is the normalization factor, which makes the result of the summation term fall within a reasonable numerical range; is the weighting coefficient; P(t) is the real-time transmission power of the line; This represents the maximum allowable transmission power of the line. >1 indicates that the network structure is in a highly vulnerable state; To determine the load transfer sensitivity, calculate the load transfer rate of other lines after the critical line is interrupted. High sensitivity means that a fault is likely to trigger a cascading overload.

[0126] Equation (7)

[0127] In the formula, The number of other lines that are electrically connected to line l; The power increment of line l after line k is disconnected; This represents the power transmitted before line k was disconnected. With a value close to 0, this line is largely unaffected by faults in other lines. Approaching 1: When other lines fail, almost all power flow is transferred to this line;

[0128] To protect against vulnerability, it is used to assess the probability of protection mismatch or over-tripping based on information such as current settings and switch status.

[0129] Equation (8)

[0130] In the formula, , These are the weighting coefficients. , These are the operating currents of the upstream and downstream line protection, respectively. To protect uncovered areas, For the entire region;

[0131] Step S23: Based on the calculated basic risk value, coupling correction coefficient, network structure risk coefficient, and dynamic weights calculated by combining the matching degree of the fault precursor feature library, generate dynamic weights and calculate the fault transfer risk index. The calculation formula is shown below:

[0132] Equation (9)

[0133] In the formula, The weighting is dynamic, adjusted based on the matching degree between real-time indicators and the fault precursor feature library. The fault precursor feature library is established from historical cases or simulation analysis, defining typical change patterns of each indicator before a specific fault transfer mode occurs (e.g., a surge in partial discharge pulse frequency accompanied by a specific harmonic increase); matching degree... High indicators have high weight. Significantly improved in a short period of time:

[0134] Equation (10)

[0135] in, For static base weights, To enhance the model's ability to respond hypersensitively to signs that it has entered a specific failure evolution path.

[0136] In one embodiment, step S3: Based on the value of the failover risk index, trigger a tiered early warning and generate corresponding handling decision recommendations, specifically including the following steps:

[0137] Step S31: The system compares the threshold value based on the dynamic failover risk index FTR(t), automatically determines and triggers the corresponding level of warning.

[0138] Specifically, the system has the following three-level early warning mechanism preset:

[0139] When 0.3≤FTR(t)<0.5, a blue alert (attention level) is triggered, indicating that risk indicators have appeared but a clear failover path has not yet been formed;

[0140] When 0.5≤FTR(t)<0.7, a yellow warning (risk level) is triggered, indicating that the indicator combination has matched the historical fault precursor pattern and the possibility of fault transfer is high.

[0141] When FTR(t)≥0.7, a red alert (emergency level) is triggered, indicating that the system is in a highly vulnerable state and multiple strongly coupled risk indicators show that failover is imminent.

[0142] Step S32: Based on the triggered warning level, the system automatically generates a structured warning report containing the main risk indicators, precise location, and tiered handling suggestions.

[0143] Furthermore, the report will first identify the leading indicators that contribute the most to the current risks and their real-time values, and then clarify the specific equipment with the highest risk and the scope of potential impact.

[0144] Subsequently, the system provides differentiated processing instructions:

[0145] In response to the blue alert, it is recommended to increase the frequency of patrols or pay attention to the trends of specific indicators;

[0146] In response to a yellow alert, specific maintenance instructions will be given, such as "conduct ultrasonic partial discharge testing within 48 hours".

[0147] In response to a red alert, a pre-decision isolation plan is automatically generated and recommended (such as suggesting the immediate disconnection of a certain communication switch) for dispatchers to make a quick decision, and an emergency plan is activated simultaneously.

[0148] Secondly, such as Figure 3 As shown, this application provides a distribution network fault transfer early warning system based on multi-dimensional fault transfer risk indicators, including:

[0149] The multi-dimensional fault transfer risk indicator perception and data fusion module 100 is used to acquire multi-source heterogeneous data from power distribution automation terminals, sensors deployed in cables and the environment, production management systems and user management systems, and perform time alignment and spatial correlation processing on the data to output panoramic status information of multi-dimensional fault transfer risk indicators.

[0150] The multi-level dynamic fusion risk assessment module 200 is communicatively connected to the multi-dimensional failover risk index perception and data fusion module 100. It is used to receive the panoramic status information, perform fusion calculation through basic risk aggregation, coupling effect correction and network structure vulnerability assessment, and output a dynamic failover risk index.

[0151] The graded early warning and response decision support module 300 is communicatively connected to the multi-level dynamic fusion risk assessment module 200. It is used to receive the fault transfer risk index, classify it according to a preset threshold, and generate response decision suggestions corresponding to the early warning level.

[0152] In one embodiment, the multi-dimensional failover risk indicator perception and data fusion module is specifically used to perform the following functions:

[0153] Real-time electrical quantity data of key nodes are collected from distribution automation terminals or smart circuit breakers through electrical quantity monitoring functions.

[0154] The device status and environment sensing functions acquire partial discharge signals, temperature and operating environment data from sensors deployed in cables and the environment, and obtain automation terminal status information from the power distribution master station system.

[0155] The system integrates offline or near real-time data from the production management system, user electricity information collection system, and marketing system through operation and maintenance and user data integration functions.

[0156] By using data fusion and modeling functions, all data is spatiotemporally aligned, digital twins of devices are constructed, and feature extraction and normalization are performed to build a panoramic status information covering the dimensions of "device-environment-operation and maintenance".

[0157] In one embodiment, the multi-level dynamic fusion risk assessment module is specifically used to perform the following functions:

[0158] The input multidimensional indicators are preprocessed to transform them into risk contribution rates with unified dimensions.

[0159] The basic risk aggregation function can be used to aggregate similar independent risk indicators and add them together to form a system-level basic risk vector.

[0160] By using the coupling effect correction function, the interaction between different categories of risk indicators is quantified, and the coupling correction coefficient is calculated to correct the underlying risk.

[0161] The network structure risk coefficient is calculated based on real-time power grid topology and power flow data using the network structure vulnerability assessment function.

[0162] Through dynamic weight adjustment and comprehensive calculation functions, the risk weights are dynamically adjusted according to the matching degree between real-time indicators and historical fault precursor feature database, and the dynamic fault transfer risk index FTR(t) is obtained through comprehensive calculation.

[0163] In one embodiment, the tiered early warning and response decision support module is specifically used to perform the following functions:

[0164] The received failover risk index FTR(t) is assessed for risk level using a built-in three-level warning threshold, generating blue, yellow, or red warnings. Based on the assessed risk level, a structured warning report is automatically generated, which includes the risk level, index value, dominant risk indicator, risk location information, and possible failover paths. Differentiated handling decision recommendations are generated in conjunction with the warning level, including inspection recommendations for blue warnings, targeted operation instructions for yellow warnings, and pre-decision isolation plans for red warnings.

[0165] In one embodiment, the system further includes a data storage and management module, a human-machine interface, and an external communication interface. The data storage and management module is used to store historical data, model parameters, feature libraries, and early warning records; the human-machine interface is used to visually display the overall status, risk index, early warning information, and handling suggestions; the external communication interface is used to interact with scheduling automation systems, production management systems, etc., and issue commands, thereby forming a complete closed loop from risk perception, dynamic assessment, intelligent early warning to decision support.

[0166] The functions of each module in the above-mentioned distribution network transfer early warning system based on multidimensional fault transfer risk indicators correspond to the steps in the above-mentioned distribution network transfer early warning method embodiment based on multidimensional fault transfer risk indicators. Their functions and implementation processes will not be described in detail here.

[0167] Thirdly, this application provides a distribution network transfer early warning device based on a multi-dimensional fault transfer risk index. The distribution network transfer early warning device based on the multi-dimensional fault transfer risk index can be a personal computer (PC), a laptop, a server, or other device with data processing capabilities.

[0168] In this embodiment of the application, the distribution network transfer early warning device based on multi-dimensional fault transfer risk indicators may include a processor, a memory, a communication interface, and a communication bus.

[0169] The communication bus can be of any type and is used to interconnect the processor, memory, and communication interface.

[0170] The communication interface includes input / output (I / O) interfaces, physical interfaces, and logical interfaces. These interfaces enable interconnection of devices within the distribution network transfer early warning equipment based on multi-dimensional fault transfer risk indicators, and also enable interconnection between the equipment and other devices (such as other computing devices or user equipment). Physical interfaces can be Ethernet interfaces, fiber optic interfaces, ATM interfaces, etc.; user equipment can be displays, keyboards, etc.

[0171] Memory can be various types of storage media, such as random access memory (RAM), read-only memory (ROM), non-volatile RAM (NVRAM), flash memory, optical storage, hard disk, programmable ROM (PROM), erasable PROM (EPROM), electrically erasable PROM (EEPROM), etc.

[0172] The processor can be a general-purpose processor, which can call a distribution network transfer early warning program based on multi-dimensional fault transfer risk indicators stored in memory and execute the distribution network transfer early warning method based on multi-dimensional fault transfer risk indicators provided in the embodiments of this application. For example, the general-purpose processor can be a central processing unit (CPU). The method executed when the distribution network transfer early warning program based on multi-dimensional fault transfer risk indicators is called can refer to the various embodiments of the distribution network transfer early warning method based on multi-dimensional fault transfer risk indicators in this application, and will not be repeated here.

[0173] Fourthly, embodiments of this application also provide a readable storage medium.

[0174] The present application stores a distribution network transfer early warning program based on a multi-dimensional fault transfer risk index on a readable storage medium. When the distribution network transfer early warning program based on the multi-dimensional fault transfer risk index is executed by a processor, it implements the steps of the distribution network transfer early warning method based on the multi-dimensional fault transfer risk index as described above.

[0175] The method implemented when the distribution network transfer early warning program based on multidimensional fault transfer risk indicators is executed can be referred to in various embodiments of the distribution network transfer early warning method based on multidimensional fault transfer risk indicators in this application, and will not be repeated here.

[0176] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators, characterized in that, Includes the following steps: Acquire and integrate multi-dimensional fault transfer risk indicator data of the distribution network to form a panoramic status information; Based on the panoramic status information, a dynamic failover risk index is calculated using a multi-level fusion evaluation model. Based on the value of the failover risk index, a tiered early warning is triggered and corresponding handling decision recommendations are generated.

2. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 1, characterized in that, The process of acquiring and integrating multi-dimensional fault transfer risk index data of the distribution network to form panoramic status information specifically includes the following steps: Collect real-time electrical quantity data, equipment status data, and operating environment data of the power distribution network; Obtain operation and maintenance management data of the power distribution network and user-side status data; By performing time alignment and spatial correlation on the above multi-source heterogeneous data, a panoramic status information of multi-dimensional fault transfer risk indicators under a unified spatiotemporal benchmark is constructed.

3. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 2, characterized in that, The equipment status data collected, which includes the real-time electrical quantity data, equipment status data, and operating environment data of the power distribution network, includes: The functional operation status and setting information of the power distribution automation terminal are obtained; Sensors deployed at cable joints are used to monitor partial discharge signals and joint temperature online.

4. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 1, characterized in that, The calculation of the dynamic failover risk index based on the panoramic state information and through a multi-level fusion evaluation model specifically includes the following steps: In the first basic risk aggregation layer, the standardized risk values ​​of each independent failover risk indicator are calculated and aggregated to obtain the basic risk value; In the second coupling effect correction layer, the coupling correction coefficients for the interaction between different categories of failover risk indicators are calculated; In the third network structure vulnerability assessment layer, the network structure risk coefficient is calculated based on real-time power grid topology and power flow. By combining the basic risk value, coupling correction coefficient, and network structure risk coefficient, a dynamic failover risk index is calculated.

5. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 4, characterized in that, The failover risk index The calculation formula is shown below: ； in, For the first The dynamic weight of each basic risk. Its basic risk value, In order to be with the first The product of all coupling correction coefficients related to the basic risk, This represents the network structure risk coefficient.

6. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 4, characterized in that, The calculation of the standardized risk value includes: For continuously monitored time-domain and frequency-domain indicators, a hybrid normalization method based on threshold and statistical distribution is used for conversion; For discontinuous event-based status indicators, convert them into risk values ​​with a time-decrease factor.

7. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 4, characterized in that, A dynamic weight adjustment mechanism is introduced in the calculation of the dynamic failover risk index. The improvement is based on the matching degree between real-time indicator data and a pre-built fault precursor feature library, and the calculation formula is as follows: ； in, For static base weights, As an enhancing factor, For real-time matching accuracy.

8. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 1, characterized in that, The tiered early warning system is a three-level early warning system, including: When the failover risk index is greater than or equal to the first preset threshold and less than the second preset threshold, a first-level warning is triggered and an inspection suggestion is generated. When the failover risk index is greater than or equal to the second preset threshold and less than the third preset threshold, a second-level warning is triggered, and a warning report containing the dominant risk indicators and targeted handling suggestions is generated. When the failover risk index is greater than or equal to the third preset threshold, a third-level warning is triggered, and alarm information and a pre-decision isolation scheme are generated. The first preset threshold, the second preset threshold, and the third preset threshold increase sequentially.

9. The distribution network fault transfer early warning method based on multi-dimensional fault transfer risk indicators according to claim 8, characterized in that, The recommended actions in the Level 2 warning report include conducting special testing or verifying protection settings within a specified timeframe.

10. A distribution network fault transfer early warning system based on multi-dimensional fault transfer risk indicators, characterized in that, include: The multi-dimensional fault transfer risk indicator perception and data fusion module is used to acquire multi-source heterogeneous data from power distribution automation terminals, sensors deployed on cables and in the environment, production management systems and user management systems, and to perform time alignment and spatial correlation processing on the data to output panoramic status information of multi-dimensional fault transfer risk indicators. The multi-level dynamic fusion risk assessment module is communicatively connected to the multi-dimensional failover risk index perception and data fusion module. It is used to receive the panoramic state information, perform fusion calculations through basic risk aggregation, coupling effect correction and network structure vulnerability assessment, and output a dynamic failover risk index. The graded early warning and response decision support module is communicatively connected to the multi-level dynamic fusion risk assessment module. It is used to receive the fault transfer risk index, classify it according to a preset threshold, and generate response decision suggestions corresponding to the early warning level.

Images