CN-120999592-B - Power distribution network fault early warning method and system based on online monitoring

CN120999592BCN 120999592 BCN120999592 BCN 120999592BCN-120999592-B

Abstract

The application discloses a power distribution network fault early warning method and system based on-line monitoring, and relates to the field of fault early warning; the method comprises the steps of completing signal waveform prediction of a real-time electric signal according to a real-time signal prediction result, judging whether an early fault signal exists in a target power distribution network, acquiring power distribution network information, environment prediction data and geographic position data if the early fault signal exists in the target power distribution network, constructing a power distribution network topological structure, extracting the early fault information of the target power distribution network, constructing an icing evolution model of the target power distribution network, completing fault risk coupling between the icing evolution model and the early fault information, performing risk accumulation analysis on the power distribution network topological structure, and outputting fault early warning information of the target power distribution network according to a risk accumulation analysis result. The application can effectively improve the fault early warning precision and efficiency of the power distribution network.

Inventors

LI JUN
CHENG XI
WANG KANG
LI PANPAN
GAN CHENGPENG
LUO GANG
GE YANPING
CHEN JIANHUI
YANG JIAN
HE XIAOHONG
FENG CAIBO
WANG JIANPING
CAO ZHENGQIANG
LIANG JUNXIONG

Assignees

湖北网安科技有限公司

Dates

Publication Date: 20260508
Application Date: 20250808

Claims (9)

1. The power distribution network fault early warning method based on-line monitoring is characterized by comprising the following steps of: continuously acquiring real-time electrical signals of all monitoring points in a target power distribution network through a plurality of types of sensors preset in the target power distribution network; For any monitoring point, carrying out real-time signal prediction according to the real-time electric signals of the monitoring point by utilizing an extended Kalman filtering algorithm, and judging whether an early fault signal exists in the target power distribution network by combining a real-time signal prediction result and preset early fault characteristics; If an early fault signal exists in the target power distribution network, acquiring power distribution network information of the target power distribution network, and environment prediction data and geographic position data of a place where the target power distribution network is located, wherein the power distribution network information comprises line load data, and the environment prediction data comprises predicted wind directions; Constructing a power distribution network topological structure based on power distribution network information, and extracting early fault information of a target power distribution network by combining the power distribution network topological structure and all early fault signals; Determining windward characteristics of distribution lines in a target distribution network according to the trend of the distribution network lines and the predicted wind direction, wherein the windward characteristics comprise upwind lines, downwind lines and crosswind lines; Dividing a target power distribution network into a plurality of power distribution network areas according to line load data and windward characteristics of a power distribution line; preliminary building a primary icing evolution model of each power distribution network area based on environmental prediction data; For any primary icing evolution model, taking geographic position data, windward characteristics and line load data as icing correction factors, and correcting the primary icing evolution model by using the icing correction factors to obtain a regional icing evolution model; combining the ice coating evolution models of all areas to obtain an ice coating evolution model of the target power distribution network; completing fault risk coupling between the ice coating evolution model and early fault information to obtain a power distribution network risk set of a power distribution network topological structure; Repeating the steps, continuously acquiring a power distribution network risk set of the power distribution network topological structure, performing risk accumulation analysis on the power distribution network topological structure according to the continuously acquired power distribution network risk set, and outputting fault early warning information of the target power distribution network according to a risk accumulation analysis result.
2. The method according to claim 1, wherein the step of predicting the real-time signal according to the real-time electrical signal of the monitoring point by using an extended kalman filtering algorithm, and determining whether the target power distribution network has an early fault signal according to the real-time signal prediction result and a preset early fault feature comprises the following steps: synchronous sampling of real-time electrical signals of monitoring points is completed by adopting a second-order generalized integral phase-locked loop, and electrical sampling signals are obtained; according to the electrical sampling signal, completing prior estimation of the current moment of the target power distribution network by using an extended Kalman filtering algorithm to obtain a signal prediction state, wherein the signal prediction state comprises a signal prediction instantaneous value, a signal prediction frequency, a signal prediction phase and a signal prediction amplitude; Extracting a signal sampling state of an electrical sampling signal, and calculating a signal difference index between the signal sampling state and a signal prediction state; calculating the fault similarity between the signal sampling state and a preset early fault feature through a similarity formula; the early fault identification of the real-time electrical signal is completed by combining the signal difference index and the fault similarity; if the real-time electrical signal is a normal electrical signal, continuously acquiring the real-time electrical signal of the monitoring point; If the real-time electric signal is a suspected fault signal, secondary fault identification of the suspected fault signal is completed by injecting an auxiliary detection signal into the monitoring point, and whether the real-time electric signal is an early fault signal is judged according to a secondary fault identification result; and if the real-time electrical signal is an early fault signal, indicating that the target power distribution network has the early fault signal.
3. The method of claim 2, wherein the secondary fault identification of the suspected fault signal is accomplished by injecting an auxiliary detection signal into the monitoring point, and determining whether the real-time electrical signal is an early fault signal based on the secondary fault identification result comprises the steps of: The suspected fault signal is enhanced by injecting an auxiliary detection signal into the monitoring point, so that an electric enhancement signal is obtained; carrying out modal decomposition on the electric enhancement signals to obtain a plurality of signal natural modes; calculating signal modal energy of all signal natural modes by using an energy operator; Weight is distributed to all signal natural modes according to the signal mode energy, and signal reconstruction is carried out on all signal natural modes with weight distribution completed, so that an electrical reconstruction signal is obtained; Calculating the signal difference degree between the electrical reconstruction signal and a pre-acquired historical reference signal; if the signal difference is smaller than or equal to a preset difference threshold, judging that the real-time electric signal is a normal electric signal; if the signal difference is greater than the difference threshold, the real-time electrical signal is determined to be an early fault signal.
4. The method of claim 1, wherein the power distribution network information further comprises power distribution network structure data, and the environmental prediction data further comprises predicted temperature, predicted humidity, predicted wind speed, predicted precipitation, and predicted time to freezing rain.
5. The method according to claim 4, wherein the constructing the power distribution network topology based on the power distribution network information and extracting the early fault information of the target power distribution network by combining the power distribution network topology and all the early fault signals comprises the steps of: positioning early fault positions of all early fault signals by using a signal positioning algorithm; Abstracting a target power distribution network into a power distribution network topological structure according to power distribution network structure data, wherein nodes of the power distribution network topological structure comprise power supply nodes, branch nodes, monitoring nodes, equipment nodes and fault nodes, and edges of the power distribution network topological structure are power distribution lines of the target power distribution network; For any early-stage fault signal, extracting topological fault characteristics of the early-stage fault signal based on a power distribution network topological structure, wherein the topological fault characteristics comprise fault branch coefficients, fault reflection coefficients and fault load coefficients; extracting basic fault characteristics of early fault signals based on a wavelet transformation algorithm; Stacking fault feature tensors by combining topology fault features and basic fault features; inputting the fault characteristic tensor into a pre-constructed signal fault recognition model, outputting the early fault type of the target power distribution network through the fault signal recognition model, and constructing the signal fault recognition model based on a residual error network; And integrating the early fault position and the early fault type of the early fault signal into the early fault information of the target power distribution network.
6. The method of claim 5, wherein the extracting the base fault signature of the early fault signal based on the wavelet transform algorithm comprises the steps of: preprocessing an early failure signal; Performing stable wavelet transformation decomposition on the early failure signals subjected to pretreatment to obtain a plurality of early failure subbands; equally dividing each early failure sub-band into a plurality of early failure segments; And extracting basic fault characteristics of all early fault sections, wherein the basic fault characteristics comprise a fault mean value, a fault variance, a fault energy value, a fault energy entropy, a fault frequency entropy value and a fault entropy weight.
7. The method of claim 4, wherein the completing the fault risk coupling between the ice-over evolution model and the early fault information to obtain a power distribution network risk set for the power distribution network topology comprises the steps of: for any fault node in the power distribution network topological structure, determining a fault evolution grade and a fault risk grade of the fault node according to an early fault type in early fault information corresponding to the fault node, and distributing an early fault risk value for the fault node by combining the fault evolution grade and the fault risk grade; calculating the line ice coating thickness at the position corresponding to the fault node according to the ice coating evolution model; Determining an icing type of the target power distribution network by combining the predicted temperature, the predicted humidity and the predicted wind speed, wherein the icing type comprises stable icing and unstable icing; Distributing a line icing risk value for the fault node by combining the line icing thickness and the icing type; determining the topology centrality of fault nodes according to the topology structure of the power distribution network, and distributing fault topology risk values for the fault nodes according to the topology centrality; weighting and fusing the early fault risk value, the line icing risk value and the fault topology risk value to obtain a node coupling risk value of the fault node; and integrating node coupling risk values of all fault nodes to obtain a power distribution network risk set of the power distribution network topological structure.
8. The method according to claim 1, wherein the step of performing risk accumulation analysis on the power distribution network topology according to the continuously acquired power distribution network risk set and outputting fault early warning information of the target power distribution network according to the result of the risk accumulation analysis comprises the following steps: Performing time integration on all fault nodes in a power distribution network topological structure according to the continuously acquired power distribution network risk set to obtain node accumulated risk values of all fault nodes; and outputting fault early warning information of the target power distribution network when the node accumulated risk value of any fault node is greater than or equal to a preset risk threshold value.
9. An on-line monitoring-based power distribution network fault early warning system is characterized by comprising: A memory configured to store instructions, and A processor configured to invoke instructions from a memory and when executing the instructions is capable of implementing the on-line monitoring based power distribution network fault pre-warning method according to any one of claims 1 to 8.

Description

Power distribution network fault early warning method and system based on online monitoring Technical Field The embodiment of the application relates to the field of fault early warning, in particular to a power distribution network fault early warning method and system based on-line monitoring. Background The power distribution network is used as a key link for connecting a power transmission network and a user in a power system, and the operation stability of the power distribution network is directly related to the reliability and the safety of power supply. However, the distribution network has wide distribution range and complex topological structure, and is easily influenced by multiple factors such as natural environments (such as ice coating, lightning stroke, typhoon and the like), equipment aging, load fluctuation and the like. Particularly, under extreme natural conditions, such as snowstorm weather, typhoon weather and the like, the probability of the occurrence of faults of the power distribution network is greatly increased, so that the power distribution network is required to be monitored in real time and is subjected to fault early warning, and the large-scale power failure event caused by the expansion of the fault range is prevented. The existing power distribution network fault early warning technology mainly relies on single-type or multi-type sensor data, and the power distribution network fault is simply identified by signal analysis of the sensor data, so that a tiny fault signal is difficult to identify by the method, and fault missed judgment possibly occurs. In addition, single signal fault analysis is carried out on the power distribution network, and other factor cooperative analysis is lacked, so that error can occur in fault detection results, the fault information of the power distribution network is lagged, and the requirements of refined and intelligent operation and maintenance of the power distribution network are difficult to meet. Disclosure of Invention The embodiment of the application provides a power distribution network fault early warning method and system based on-line monitoring, which are used for solving the problem that the power distribution network fault is difficult to accurately and timely identify in the prior art. In order to achieve the above purpose, the embodiment of the present application adopts the following technical scheme: In a first aspect, a power distribution network fault early warning method based on online monitoring is provided, and the method includes: continuously acquiring real-time electrical signals of all monitoring points in a target power distribution network through a plurality of types of sensors preset in the target power distribution network; For any monitoring point, carrying out real-time signal prediction according to the real-time electric signals of the monitoring point by utilizing an extended Kalman filtering algorithm, and judging whether an early fault signal exists in the target power distribution network by combining a real-time signal prediction result and preset early fault characteristics; If the target power distribution network has an early fault signal, acquiring power distribution network information of the target power distribution network, and environment prediction data and geographic position data of a place where the target power distribution network is located; Constructing a power distribution network topological structure based on power distribution network information, and extracting early fault information of a target power distribution network by combining the power distribution network topological structure and all early fault signals; constructing an ice coating evolution model of the target power distribution network by combining power distribution network information, environment prediction data and geographic position data; completing fault risk coupling between the ice coating evolution model and early fault information to obtain a power distribution network risk set of a power distribution network topological structure; Repeating the steps, continuously acquiring a power distribution network risk set of the power distribution network topological structure, performing risk accumulation analysis on the power distribution network topological structure according to the continuously acquired power distribution network risk set, and outputting fault early warning information of the target power distribution network according to a risk accumulation analysis result. Optionally, the method for predicting the real-time signal according to the real-time electric signal of the monitoring point and by using the extended kalman filtering algorithm, and judging whether the target power distribution network has the early fault signal by combining the real-time signal prediction result and the preset early fault feature comprises the following steps: synchronous sampling of real-time electrical signals of m