CN-114417732-B - Multi-source load damage self-adaptive identification method and system for power distribution network under strong typhoon

Abstract

The invention provides a multi-source load damage self-adaptive identification method and system for a power distribution network under strong typhoons, wherein the method comprises the steps of carrying out fault judgment on node communication of the power distribution network under the strong typhoons, carrying out multi-source load power distribution network topology identification by using collected node electric quantity information of the power distribution network if no communication fault exists, obtaining fault types and fault positions, then obtaining node loss probability through disaster damage identification correction, obtaining multi-dimensional original meteorological information and node information if communication faults exist, correcting wind speeds in the multi-dimensional original meteorological information by considering ground roughness and relative heights, obtaining corrected data sequences, inputting the corrected data sequences into a deep learning disaster damage prediction model, obtaining post-disaster conditions and node element damage conditions for a certain time in the future, and then obtaining independent loss probability of single nodes of the multi-source load of the power distribution network by using power distribution network topology identification.

Inventors

• LI GUANGXIAO
• WEI YUANYUAN
• LIU YING
• SHAO SHIWEN
• WANG XIAOYE
• NI XINXIN
• YANG ZHIPENG
• MENG YANGYANG
• LIU HUALI
• ZHAO MENG
• WANG LIN
• LIU SIXIAN
• HE ZHAOHUI
• LIU ZONGJIE
• TIAN CHONGFENG
• SUN WENSHENG
• SHI ZHIGUO
• WU DONG

Assignees

• 国网山东省电力公司济宁供电公司
• 国家电网有限公司

Dates

Publication Date

20260505

Application Date

20220128

Claims (9)

1. 1. The method for adaptively identifying the multi-source load damage of the power distribution network under the strong typhoons is characterized by comprising the steps of performing fault judgment on node communication of the power distribution network under the strong typhoons; If no communication fault exists, carrying out topology identification on the multi-source load distribution network by using the collected node electric quantity information of the distribution network to obtain a fault type and a fault position, and then obtaining the node power failure probability through disaster damage identification and correction; If communication faults exist, multi-dimensional original meteorological information and node information are obtained, wind speed in the multi-dimensional original meteorological information is corrected by considering ground roughness and relative height, a corrected data sequence is obtained, the corrected data sequence is input into a deep learning disaster damage prediction model, a disaster post-condition and a node element damage condition of a certain time in the future are obtained, and then independent power failure probability of a single node of a multi-element source load of a power distribution network is obtained by utilizing power distribution network topology identification; The wind speed in the multidimensional original meteorological information is corrected by considering the roughness and the relative height of the ground, and the method specifically comprises the following steps: let the reference wind speed measured by the nearby weather station be The surface roughness is Unified height h meters for measuring wind, and corrected wind speed of nodes Is that The surface roughness is ; The wind speed change caused by the relative altitude change is that the relative altitude of the node and the reference station is set as Wind speed caused by terrain Is changed into : Wherein A, B is a topographic parameter, L is the length of the power line, Representing the relative wind speed to a reference due to a change in terrain elevation Δh Is a ratio of wind speed variation; the change of wind speed caused by the roughness change of the ground is set as The formula is calculated: , wherein, Is the height of the boundary layer of the atmosphere, also known as the top of the surface roughness layer; Thus the wind speed after node correction Expressed by the following formula: 。
2. 2. The method for adaptively identifying the multi-source load damage of the distribution network under the strong typhoons as set forth in claim 1, wherein the topology identification of the multi-source load distribution network specifically includes: acquiring the voltage, current and load of each node in the power distribution network in real time; Defining a node state matrix S of the power distribution network, wherein the S is a1 Xn-order matrix, n elements are 1 before a disaster occurs, and n nodes are in a normal power supply state; Each node in the power distribution network is numbered, a node connection relation matrix is defined, a fault matrix F is defined, and fault location and cause analysis are carried out.
3. 3. The method for adaptively identifying the multi-source load and disaster damage of the power distribution network under strong typhoons according to claim 1, wherein the fault matrix F is defined, fault location and cause analysis are performed, and the method comprises the following steps: when the fault matrix F is assigned: the k node is powered off, and the k-1 node and the k+1 node are normal, so that the transformer at the k node is indicated to be faulty; searching an upper node number of k and a lower node number of k by using the connection relation matrix, and then obtaining the value of the (k, k) th coordinate in the fault matrix F; defining a fault matrix F, carrying out fault positioning and cause analysis, and further comprising a second case: The k node is powered off, the k-1 node is normal, and the k+1 node is powered off; Traversing all lower nodes of the branch where the k is located, if the power is lost, breaking line faults occur between the k and the k-1, and if the number of normal lower nodes is not 0, the transformers at the k and the k+1 are indicated to be simultaneously broken.
4. 4. The method for adaptively identifying the multi-source load damage of the distribution network under the strong typhoons as set forth in claim 3, wherein a fault matrix F is defined, fault location and cause analysis are carried out, and the method further comprises the following steps: The kth node is powered off, and the k-1 node and the k+1 node are powered off; Suppose that the upper node of k-1 fails and needs to jump to the specific analysis of the node; And then, circularly judging the execution condition I and the execution condition II until nodes corresponding to all fault values in the power distribution network node state matrix S are traversed once, and finally reflecting the fault positions and the fault types according to the values in the matrix F.
5. 5. The method for adaptively identifying the multi-source load loss of the power distribution network under the strong typhoons according to claim 1, wherein the node power failure probability is obtained through disaster loss identification and correction, and the method specifically comprises the following steps: considering the back-off bar and the wire breakage model, let its probability be respectively And The independent power loss probability of the single node of the multi-source charge of the power distribution network is as follows: 。
6. 6. The method for adaptively identifying the multi-source load damage of the distribution network under the strong typhoons according to claim 1, wherein a deep learning network is trained according to weather and disaster damage history data to obtain a deep learning disaster damage prediction model for inputting weather and outputting disaster damage.
7. 7. Multi-source load damage self-adaptive identification system of power distribution network under strong typhoons is characterized by comprising: The communication fault judging module is configured to judge the fault of node communication of the power distribution network under strong typhoons; The first node power failure probability calculation module is configured to perform multi-source load distribution network topology identification by using the collected node electric quantity information of the distribution network if no communication fault exists, obtain a fault type and a fault position, and obtain node power failure probability through disaster damage identification and correction; The second node power failure probability calculation module is configured to obtain multi-dimensional original meteorological information and node information if communication faults exist, correct wind speed in the multi-dimensional original meteorological information by considering ground roughness and relative height, obtain a corrected data sequence, input the corrected data sequence into a deep learning disaster damage prediction model to obtain post-disaster conditions and node element damage conditions of a certain time in the future, and then obtain independent power failure probability of a single node of a multi-element source load of the power distribution network by using power distribution network topology identification, wherein the wind speed in the multi-dimensional original meteorological information is corrected by considering ground roughness and relative height, and specifically comprises the following steps: let the reference wind speed measured by the nearby weather station be The surface roughness is The wind measuring unified height is h meters, and the corrected wind speed of the nodes is The surface roughness is ; The wind speed change caused by the relative altitude change is that the relative altitude of the node and the reference station is set as Wind speed caused by terrain Is changed into : Wherein A, B is a topographic parameter, L is the length of the power line, Representing the relative wind speed to a reference due to a change in terrain elevation Δh Is a ratio of wind speed variation; the change of wind speed caused by the roughness change of the ground is set as The formula is calculated: , wherein, Is the height of the boundary layer of the atmosphere, also known as the top of the surface roughness layer; Thus the wind speed after node correction Expressed by the following formula: 。
8. 8. A computing device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method of any of the preceding claims 1-6 when the program is executed.
9. 9. A computer readable storage medium, on which a computer program is stored, characterized in that the program, when being executed by a processor, performs the steps of the method of any of the preceding claims 1-6.

Description

Multi-source load damage self-adaptive identification method and system for power distribution network under strong typhoon Technical Field The invention belongs to the technical field of disaster damage identification of power distribution networks, and particularly relates to a method and a system for adaptively identifying multi-source load damage of a power distribution network under strong typhoons. Background The statements in this section merely provide background information related to the present disclosure and may not necessarily constitute prior art. The large-scale power failure caused by frequent typhoon disasters brings great challenges to the toughness of a power distribution network with the characteristics of multiple voltage levels, complex network structure, multiple equipment types and the like, and greatly hinders the important task of providing electric power energy for various users. Current research on fault recovery strategies under extreme disasters usually solves the optimal fault recovery strategy through different optimization algorithms on the basis of fault location and fault isolation. Therefore, if topology identification and fault location are carried out on high-risk areas and ultra-short-term power loss loads in the disaster through disaster damage modeling and deep learning algorithms, the maximum recovery of the loads can be carried out in the least time, and casualties and economic losses are reduced. (1) The method for predicting the power distribution network large-scale outage probability under the extreme meteorological conditions is provided by a national key laboratory Chen Ying of a motor system electric power system of Qinghai university, and the method for modeling the power distribution network fault probability comprises a power distribution network equipment outage event Bayesian network model under typhoon disasters and node outage probability prediction based on the power distribution network equipment outage event Bayesian network. The method overcomes the defect that the original research is difficult to describe the dynamic characteristics of the weather, and fully describes the causal dependency relationship and the uncertainty in the power distribution network power failure event by using the Bayesian network. According to the method, only data driving is used as a basis, the space-time correlation of equipment outage events after disaster is considered, historical disaster damage records and disaster numerical simulation data are utilized to construct a disaster time Bayesian network model, and then the power outage range and the power outage probability of the power distribution network are rapidly inferred according to disaster conditions. (2) The national key laboratory Wang Zeng of new energy power system of North China electric power system, ping et al, propose a 110kV line reverse tower and disconnection accident assessment method under typhoon and storm disaster. The method comprises the step of analyzing the influence of typhoon storm models and typhoon storm on 110kV conductors and iron towers. The method comprises the steps of respectively establishing finite element models for different power transmission towers of 110kV lines, analyzing power responses of the lines and towers under wind and rain loads by considering the correlation of the wind and rain loads, deducing broken lines and inverted tower probability expressions based on a structural reliability theory, researching the mechanism influence of factors such as different towers, purposes, wind direction angles and the like on inverted towers, analyzing weak links of a power grid, and evaluating inverted towers and broken line accidents of the 110kV lines under typhoon and storm disasters. The method is based on mechanism analysis only, derives an element fault probability expression based on a structural reliability theory, and analyzes the weak point of the power grid. (3) Fu Jianfeng of the university of delftiro proposes an unmanned aerial vehicle real-time routing strategy for monitoring and checking the recovery of a power distribution network after disaster. The method comprises the steps of monitoring recovery after disaster of the power distribution network and checking a coordinated unmanned aerial vehicle real-time routing strategy. Through the proposed real-time unmanned aerial vehicle routing strategy, the unmanned aerial vehicle can check the damage condition so as to recover after disaster. In addition, the transmission line may be monitored for potential hazards, and the road infrastructure may also be monitored to provide real-time information about traffic conditions so that maintenance personnel may choose the best way to reach the damaged area. The method combines a data driving and mechanism analysis method, and monitors damage conditions through a real-time unmanned aerial vehicle routing strategy so as to recover after disaster. In short, in the method (1), o