CN108594076A - A kind of power distribution network power-off fault analysis method - Google Patents

Abstract

The invention discloses a method for researching and judging a power outage fault in a distribution network. Overall path matrix; judge the distribution of ant colony to determine the sub-path matrix of ant colony search, and use it as a sub-pheromone matrix; use ant colony algorithm to search for faults, search for paths for all ants in the ant colony, and update them synchronously The element value of the sub-pheromone matrix; all ants have been judged as the termination condition, and the maximum value of the element in the sub-pheromone matrix is the power failure point. The invention improves the speed and accuracy of fault location, shortens the response time, reduces the cost, and improves customer satisfaction and system operation indicators without increasing hardware investment cost.

Description

A method for researching and judging power outage faults in distribution network

technical field

The invention relates to a method for researching and judging a power outage fault of a distribution network, and belongs to the technical field of distribution networks.

Background technique

Since the introduction of the smart grid in 2009, my country has comprehensively promoted the construction of smart distribution networks, and the automation level of the grid has been continuously improved. At the same time, users have raised higher requirements for the reliability and quality of power supply services. Positioning has also become a hot spot of research attention.

For a long time, for low-voltage side fault information, the power supply company has mainly used the repair calls of residents, used the power outage management system to handle low-voltage lying network faults, and predicted the location of the drop insurance or switch based on the calls made by users after the power outage. Analyze the personnel strength and emergency repair plan, determine the priority of emergency repair, calculate the work force required on site, estimate the recovery time, and manage the on-site work.

The power outage management system is a subsystem of the distribution network management system. The control room relies on this system to organize and dispatch power outage repairs and dispatch work orders. The fault handling method using the power outage management system has the following shortcomings: first, the accuracy of judgment needs to be improved; Or cables to achieve low-voltage power distribution, but there are many low-voltage users and the lines are complicated, so it is difficult to find after the line fails.

At present, the power consumption information acquisition system (AMI) and grid geographic information system (GIS) have been constructed and fully covered in the distribution network. The use of the AMI system and GIS system will help the rapid analysis and judgment of power outages in the distribution network.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies in the prior art, provide a method for researching and judging the power outage fault of the distribution network, and solve the technical problems in the prior art such as difficulty in finding the fault point of the power outage in the distribution network, low efficiency, and long time consumption.

In order to solve the above-mentioned technical problems, the technical solution adopted in the present invention is: 1. A method for researching and judging a power outage fault in a distribution network, characterized in that, the method includes the following steps:

Trigger the function of power outage fault research and judgment, and determine the target range of calling and testing;

Screen out power outage and voltage loss users and obtain the distribution network topology of the power outage and voltage loss users;

Ant colonies are established for power outage and voltage loss users, and an overall path matrix is established according to the topology of the distribution network;

Judging the ant colony distribution to determine the sub-path matrix of the ant colony search, which is used as the sub-pheromone matrix;

Use the ant colony algorithm to search for faults, search for paths for all ants in the ant colony, and update the element values of the sub-pheromone matrix synchronously;

Assuming that all ants have been judged as the termination condition, the maximum value of elements in the sub-pheromone matrix is the power failure point.

Further, the method for triggering the power failure fault research and judgment function includes: automatic triggering of abnormal voltage monitoring and active triggering of power failure alarm.

Further, the automatic triggering condition of the voltage monitoring abnormality is: a single electric meter collects a voltage value of 0V for two consecutive times, or the voltage value of two or more users at the same time is 0V.

Further, the active triggering condition of the power outage alarm is: the power outage management system receives a power outage repair call from a user.

Further, the method for determining the target scope of the survey is as follows:

Step 101) judge and trigger the 10kV/400V low-voltage transformer where the user is located, and directly call and measure the coverage area of the low-voltage transformer;

Step 102) calling and measuring at least two electric meters at each branch box;

103) During the research and judgment process, continue to judge, and if user faults outside the call-and-test area continue to occur, repeat steps 101) and 102) in the new area.

Further, the method of screening power outage and voltage loss users is as follows:

The AMI system is used to judge the user voltage data: if the user voltage is 0, the user is judged as a power failure and voltage loss user, and the asset number, node number, and distribution network location information of the power failure and voltage loss user are further obtained; If the voltage is normal, it is discarded, and the next user is screened and judged.

Further, the method to obtain the topology of the distribution network for power outage and voltage loss users is as follows:

Obtain the distribution network topology where the power outage and voltage loss users are located through the GIS system connected with the power outage and voltage loss users, including: obtaining the node number of the distribution network and the user node label.

Further, the specific method of establishing an ant colony is as follows:

According to the location of power outage and voltage loss users, they are classified and aggregated to form large and small ant colonies with containment relations. From small to large, they are: small ant colonies of the same branch line, ant colonies of the same low-voltage busbar, ant colonies of the same medium-voltage feeder line and unified central Press the busbar ant colony.

Further, the specific method of establishing the overall path matrix is as follows:

Code the nodes of the 10kV medium-voltage distribution network and the 400V low-voltage distribution network separately;

Assuming that the overall path matrix is P, the overall path matrix P is split into the medium-voltage grid path matrix P _H and the low-voltage grid path matrix P _L , and the path matrices are respectively established:

Construct the path matrix P _H of the medium-voltage power grid, assuming that the total number of nodes is n, all nodes are arranged in order horizontally and vertically, and the elements of the matrix represent the path between two nodes, and the subscripts of the elements are the node numbers at both ends of the path :

Construct the path matrix _PL of the low-voltage power grid, assuming that the total number of nodes is m, and all nodes are arranged in order horizontally and vertically. The elements of the matrix represent the path between two nodes, and the element subscripts are the node numbers at both ends of the path:

Further, the method of establishing the sub-pheromone matrix is as follows:

Step 201) Determine the voltage level of the ant colony. If all the ant colonies are 400V low-voltage users, then judge whether all the ant colonies in the ant colony are in the same low-voltage branch line: if so, trace the research and judgment range back to the root node of the branch line, and list The node numbers of the root node and downstream nodes form a sub-pheromone matrix on the path matrix _PL of the low-voltage power grid, that is, the actually judged pheromone matrix P _A , where A is the number of the fault point to be searched; otherwise, go to the next step;

Step 202) Judging whether all the ant colonies in the ant colony are in the same low-voltage busbar: if so, then trace the scope of research and judgment back to the root node of the low-voltage busbar, list the node numbers of the root node and downstream nodes, and form a child on the low-voltage grid path matrix _PL Pheromone matrix, that is, the actual judged pheromone matrix P _B , B is the number of the fault point to be searched; otherwise, go to the next step;

Step 203) Determine whether all the ant colonies in the ant colony are in the same medium-voltage feeder: if so, trace the scope of research and judgment back to the root node of the feeder, list the node numbers of the root node and downstream nodes, and enter them in the medium-voltage grid path matrix P _H and low-voltage power grid path matrix _PL to form a sub-pheromone matrix, that is, the actual judged pheromone matrix P _C , C is the number of the fault point to be searched; otherwise, go to the next step;

Step 204) Judging whether all the ant colonies in the ant colony are in the same medium-voltage busbar: if so, trace the scope of research and judgment back to the root node of the busbar, list the node numbers of the root node and downstream nodes, and enter the medium-voltage grid path matrix P _H Form a sub-pheromone matrix with the path matrix _PL of the low-voltage power grid, that is, the pheromone matrix PD for actual research and judgment, and _D is the number of the fault point to be searched; if not, enter a higher-level distribution network to continue fault search and judgment;

If the user is a 10kV medium voltage user, the sub-pheromone matrix is established directly from step 203).

Further, the fault search method using the ant colony algorithm is as follows:

Step 301) The initial values of the pheromone matrix are all 0;

Step 302) Search for the target according to the number of the ant. The walking path of each ant is in the opposite direction of the trend. Walking from the initial position of the ant to the root node of the research and judgment range, all the walking paths leave pheromones, that is, in the pheromone matrix The corresponding elements of are all +1;

Step 303) all ants have completed walking, as the judgment basis of the termination condition, if there are k ant users, then k times of pheromone matrix positive iterations are performed;

Step 304) Statistically count the values of each element in the pheromone matrix, and the one with the largest value is the location of the fault point; if the element values corresponding to multiple paths in the result are equal, then it is judged that the downstream path is the fault location.

Compared with the prior art, the beneficial effects achieved by the present invention are as follows: based on the electricity information acquisition system (AMI) and the grid geographic information system (GIS) that have been constructed and fully covered by the current power distribution grid, using the ant colony algorithm, Research on the rapid location method of medium and low voltage power outages, without increasing the cost of hardware investment, improves the speed and accuracy of fault location, shortens the response time, reduces costs, and improves customer satisfaction and system operation indicators.

Description of drawings

Fig. 1 is a flow chart of the present invention;

Fig. 2 is the construction flow diagram of fault research and judgment pheromone matrix in the present invention;

Fig. 3 is a grid topological diagram of a medium and low voltage distribution network to which the present invention is applicable.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings. The following examples are only used to illustrate the technical solution of the present invention more clearly, but not to limit the protection scope of the present invention.

As shown in Figure 1, it is a flow chart of the present invention. The present invention is based on the measurement data of the AMI system, user information, and the topology data of the medium and low voltage distribution network provided by the GIS system, and uses the ant colony algorithm to perform path search and Target research and judgment includes four links: function triggering, initial data preparation, ant colony fault research and judgment process, and fault information notification forwarding. The following is a description of the four links:

Step 1: Functional triggering: Functional triggering includes automatic triggering of abnormal voltage monitoring and active triggering of power failure alarms. The automatic triggering of abnormal voltage monitoring is that a single meter collects a voltage value of 0V twice consecutively, or the voltage value of more than two users at the same time is 0V, and immediately triggers the fault research and judgment function module; the active triggering of the power failure alarm is that the power failure management system receives the user If the power outage report calls for repairs, it will actively trigger the fault research and judgment function module to find the source of the fault.

Step 2: Initial data preparation: After the power outage fault research and judgment module is triggered, it enters the initial data preparation process, including determining the target range of call-in and test, using the AMI system to conduct user voltage data call-up and screening out power outage and voltage-loss users, and obtaining call-in test by connecting with the GIS system Distribution network topology of the target range.

The method of determining the target range of the call for testing is as follows:

Step 101) Judging and triggering the 10kV/400V low-voltage transformer where the user is located, directly calling and measuring the coverage area of the low-voltage transformer; Step 102) Calling and measuring at least two electric meters at each branch box; 103) During the research and judgment process, continue to judge, if continue If there is a user fault outside the calling and testing area, repeat step 101) and step 102) in the new area.

The method of screening power outage and voltage loss users is: use AMI system to judge the user voltage data: if the user voltage is 0, then judge the user as a power outage and voltage loss user, and further obtain the asset number, node number and location of the power outage and voltage loss user The location information of the distribution network; if the user voltage is normal, it will be discarded, and the next user will be screened and judged.

The method of obtaining the topology structure of the distribution network within the target scope of the survey by connecting with the GIS system is to obtain the topology of the distribution network where the users of the power failure and voltage loss are located through the GIS system connected with the users of the power failure and voltage loss, including: obtaining the node number of the distribution network and the user node label.

Step 3: The process of ant colony fault research and judgment: firstly establish an ant colony for power failure and voltage loss users, and establish an overall path matrix according to the distribution network topology, then judge the ant colony distribution to determine the sub-path matrix for ant colony search, and use this as a sub-path matrix. The pheromone matrix, and then start to use the ant colony algorithm to search for faults, search for paths for all ants in the ant colony, and update the element values of the pheromone matrix synchronously. The maximum value is the power failure point. details as follows:

The specific method of establishing an ant colony is as follows:

According to the location of power outage and voltage loss users, they are classified and aggregated to form large and small ant colonies with containment relations. From small to large, they are: small ant colonies of the same branch line, ant colonies of the same low-voltage busbar, ant colonies of the same medium-voltage feeder line and unified central Press the busbar ant colony.

The specific method of establishing the overall path matrix is as follows:

In order to reduce the sparsity of the matrix, the nodes of the 10kV medium-voltage distribution network and the 400V low-voltage distribution network are coded separately;

Assuming that the overall path matrix is P, the overall path matrix P is split into the medium-voltage grid path matrix P _H and the low-voltage grid path matrix P _L , and the path matrices are respectively established:

Construct the path matrix P _H of the medium-voltage power grid, assuming that the total number of nodes is n, all nodes are arranged in order horizontally and vertically, and the elements of the matrix represent the path between two nodes, and the subscripts of the elements are the node numbers at both ends of the path :

Construct the path matrix _PL of the low-voltage power grid, assuming that the total number of nodes is m, and all nodes are arranged in order horizontally and vertically. The elements of the matrix represent the path between two nodes, and the element subscripts are the node numbers at both ends of the path:

As shown in Figure 2, it is the construction flowchart of the fault research and judgment pheromone matrix in the present invention, and the method for establishing the sub-pheromone matrix is:

Step 201) Determine the voltage level of the ant colony. If all the ant colonies are 400V low-voltage users, then judge whether all the ant colonies in the ant colony are in the same low-voltage branch line: if so, trace the research and judgment range back to the root node of the branch line, and list The node numbers of the root node and downstream nodes form a sub-pheromone matrix on the path matrix _PL of the low-voltage power grid, that is, the actually judged pheromone matrix P _A , where A is the number of the fault point to be searched; otherwise, go to the next step;

Step 202) Judging whether all the ant colonies in the ant colony are in the same low-voltage busbar: if so, then trace the scope of research and judgment back to the root node of the low-voltage busbar, list the node numbers of the root node and downstream nodes, and form a child on the low-voltage grid path matrix _PL Pheromone matrix, that is, the actual judged pheromone matrix P _B , B is the number of the fault point to be searched; otherwise, go to the next step;

Step 203) Determine whether all the ant colonies in the ant colony are in the same medium-voltage feeder: if so, trace the scope of research and judgment back to the root node of the feeder, list the node numbers of the root node and downstream nodes, and enter them in the medium-voltage grid path matrix P _H and low-voltage power grid path matrix _PL to form a sub-pheromone matrix, that is, the actual judged pheromone matrix P _C , C is the number of the fault point to be searched; otherwise, go to the next step;

Step 204) Judging whether all the ant colonies in the ant colony are in the same medium-voltage busbar: if so, trace the scope of research and judgment back to the root node of the busbar, list the node numbers of the root node and downstream nodes, and enter the medium-voltage grid path matrix P _H Form a sub-pheromone matrix with the path matrix _PL of the low-voltage power grid, that is, the pheromone matrix PD for actual research and judgment, and _D is the number of the fault point to be searched; if not, enter a higher-level distribution network to continue fault search and judgment;

If the user is a 10kV medium voltage user, the sub-pheromone matrix is established directly from step 203).

The method of fault search using ant colony algorithm is as follows:

Step 301) The initial values of the pheromone matrix are all 0;

Step 302) Search for the target according to the number of the ant. The walking path of each ant is in the opposite direction of the trend. Walking from the initial position of the ant to the root node of the research and judgment range, all the walking paths leave pheromones, that is, in the pheromone matrix The corresponding elements of are all +1;

Step 303) all ants have completed walking, as the judgment basis of the termination condition, if there are k ant users, then k times of pheromone matrix positive iterations are performed;

Step 304) Statistically count the values of each element in the pheromone matrix, and the one with the largest value is the location of the fault point; if the element values corresponding to multiple paths in the result are equal, then it is judged that the downstream path is the fault location.

Taking the fault at A in Figure 3 as an example, the construction steps of the pheromone matrix are as follows:

1) Judge the voltage level of the ant colony. All the ant colonies are 400V low-voltage users, and start to judge whether all the ant colonies in the ant colony are in the same low-voltage branch line;

2) Whether all the ant colonies in the ant colony are in the same low-voltage branch line, the scope of research and judgment can be traced back to the root node 3 of the branch line, and the node numbers of the root node and downstream nodes are listed. There are 7 nodes in total, and PL is formed on the basis of 7*7 sub-pheromone matrix, that is, the actual researched and judged pheromone matrix P _A , where A is the number of the fault point to be searched;

For example, the path traveled by ant user 15 is: 13-15, 10-13, 3-10, and the corresponding pheromones are increased by 1;

All ants have completed walking as the basis for judging the termination condition. If there are 4 ant users at the fault location, perform 4 positive iterations of the pheromone matrix. After the end, the pheromone matrix is:

It can be seen from the above formula pheromone matrix P _A that pa3-10 is equal to pa10-3, and both pheromones are 4. These two pheromones actually represent the same road section, that is, the line between node 3 and node 10. It can be judged that the fault location is here.

Taking the fault at B in Figure 3 as an example, the same method is used to construct it, and the brief discussion is as follows:

1) Judge the voltage level of the ant colony. If all the ant colonies are 400V low-voltage users, start to judge whether all the ant colonies in the ant colony are in the same low-voltage branch line;

2) Judging the same medium-voltage feeder, the scope of research and judgment can be traced back to the root node 6 of the feeder;

3) Construct the sub-pheromone matrix, and the research and judgment range includes 4 medium-voltage nodes and 23 low-voltage nodes, so the medium-voltage sub-pheromone matrix is a 4*4 matrix, and the low-voltage pheromone matrix is a 23*23 matrix;

4) A total of 12 ant users perform 12 forward iterations to determine the value of each element in the pheromone matrix;

5) Calculate the value of each element of the pheromone matrix, and find that the elements corresponding to the 6-7 and 7-8 paths are equal, both being 12, then judge the downstream path, that is, 7-8 is the fault location.

Step 4: Fault information notification and forwarding: After the fault location is determined, notify and forward the outage management system for the next step of investigation and power restoration.

The method of the present invention is mainly applied to medium and low voltage distribution networks. The structure of the distribution network is radial with single power supply, the flow direction is unidirectional, the medium voltage is 10kV, and the low voltage is 400V. The network topology structure is shown in FIG. 3 . Based on the electricity information acquisition system (AMI) and grid geographic information system (GIS) that have been built and fully covered by the current distribution grid, the ant colony algorithm is used to conduct research on the rapid location method of medium and low voltage power outages without adding hardware The cost of investment has improved the speed and accuracy of fault location, shortened response time, reduced costs, and improved customer satisfaction and system operation indicators.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the technical principle of the present invention, some improvements and modifications can also be made. It should also be regarded as the protection scope of the present invention.

Claims (11)

1. a kind of power distribution network power-off fault analysis method, which is characterized in that described method includes following steps：

Triggering power-off fault studies and judges function, and survey target zone is called in determination together；

Screening power failure decompression user and the distribution net topology for obtaining the power failure decompression user；

For power failure, decompression user establishes ant colony, and overall path matrix is established according to distribution net topology；

Ant colony distribution is judged to determine the subpath matrix of ant colony search, in this, as sub-information prime matrix；

Fault search is carried out using ant group algorithm, route searching, and synchronized update sub-information are carried out to all ants in ant colony The element numerical value of prime matrix；

End condition it has been judged as with all ants, element maximum value is power-off fault point in sub-information prime matrix.

2. power distribution network power-off fault analysis method according to claim 1, which is characterized in that triggering power-off fault studies and judges work( Can method include：The active triggering of the automatic trigger of voltage monitoring exception and the alarm that has a power failure.

3. power distribution network power-off fault analysis method according to claim 2, which is characterized in that the voltage monitoring exception Automatic trigger condition is：Collection voltages value is voltage value more than 2 users of 0V or same time to single ammeter twice in succession For 0V.

4. power distribution network power-off fault analysis method according to claim 2, which is characterized in that the active of the alarm that has a power failure Trigger condition is：Outage Management Systems receive the power failure repair call of user.

5. power distribution network power-off fault analysis method according to claim 1, which is characterized in that determine to call together and survey target zone Method is as follows：

Step 101) judges the 10kV/400V low-tension transformers where triggering user, directly calls together and surveys the low-tension transformer covering Region；

It is called together at each feeder pillar of step 102) and surveys at least two ammeters；

103) lasting to judge during studying and judging, if continuing to occur calling together the user malfunction surveyed outside region, repeated in new region Step 101) and step 102).

6. power distribution network power-off fault analysis method according to claim 1, which is characterized in that screening power failure decompression user's Method is as follows：

User's voltage data is judged using AMI systems：If user's voltage is 0, which is determined as power failure decompression User further obtains asset number, node serial number and the power distribution network at the place location information of power failure decompression user；Such as user Voltage is normal, then abandons, and screening judgement is carried out to next user.

7. power distribution network power-off fault analysis method according to claim 1, which is characterized in that obtain power failure decompression user and match The method of power network topology is as follows：

The distribution net topology where power failure decompression user is obtained by the generalized information system docked with power failure decompression user, including：It obtains The node serial number and user node label of power distribution network.

8. power distribution network power-off fault analysis method according to claim 1, which is characterized in that establish the specific method of ant colony It is as follows：

Grouped accumulation is carried out according to power failure decompression user position, is respectively formed the size ant colony of inclusion relation, from small to large It is respectively：Busbar ant is pressed in the small ant colony of same branch line, the ant colony of same low-voltage bus bar, same middle pressure feeder line ant colony and unification Group.

9. power distribution network power-off fault analysis method according to claim 1, which is characterized in that establish overall path matrix The specific method is as follows：

10kV medium voltage distribution networks and 400V low-voltage network nodes are encoded separately；

If overall path matrix is P, overall path matrix P is split as medium voltage network path matrix P_HWith low voltage electric network path square Battle array P_L, path matrix is established respectively：

Build medium voltage network path matrix P_HIf node total number is n, transverse and longitudinal is that all nodes arrange in order, matrix Element indicates that the path between two nodes, element subscript are the both ends node serial number in the path respectively：

Build low voltage electric network path matrix P_LIf node total number is m, transverse and longitudinal is that all nodes arrange in order, matrix Element indicates that the path between two nodes, element subscript are the both ends node serial number in the path respectively：

10. power distribution network power-off fault analysis method according to claim 9, which is characterized in that establish sub-information prime matrix Method it is as follows：

Step 201) judges the voltage class of ant colony, if ant colony is 400V low-voltage customers, judges that all ant colonies are in ant colony It is no in same low pressure branch line：If so, the root node that will study and judge range and trace back to the branch line, enumerates root node and downstream joint The node serial number of point, in low voltage electric network path matrix P_LUpper formation sub-information prime matrix, i.e., the Pheromone Matrix P actually studied and judged_A, A For the fault point to be searched number；Otherwise, into next step；

Step 202) judges that whether all ant colonies are in same low-voltage bus bar in ant colony：If so, will study and judge range traces back to low pressure The root node of busbar enumerates the node serial number of root node and downstream node, in low voltage electric network path matrix P_LUpper formation sub-information Prime matrix, i.e., the Pheromone Matrix P actually studied and judged_B, B is the fault point to be searched number；Otherwise, into next step；

Step 203) judges that whether all ant colonies are in same middle pressure feeder line in ant colony：If so, will study and judge range traces back to the feedback The root node of line enumerates the node serial number of root node and downstream node, in medium voltage network path matrix P_HWith low voltage electric network path Matrix P_LUpper formation sub-information prime matrix, i.e., the Pheromone Matrix P actually studied and judged_C, C is the fault point to be searched number；It is no Then, into next step；

Step 204) judges that whether all ant colonies are in same middle pressure busbar in ant colony：If so, will study and judge range traces back to the mother The root node of line enumerates the node serial number of root node and downstream node, in medium voltage network path matrix P_HWith low voltage electric network path Matrix P_LUpper formation sub-information prime matrix, i.e., the Pheromone Matrix P actually studied and judged_D, D is the fault point to be searched number；If It is no, continue fault search into higher level-one power distribution network and studies and judges；

If user is 10kV Middle Voltages, directly sub-information prime matrix is begun setting up from step 203).

11. power distribution network power-off fault analysis method according to claim 10, which is characterized in that carried out using ant group algorithm The method of fault search is as follows：

Step 301) Pheromone Matrix initial value is 0；

Step 302) searches for target according to the number of ant, and the walking path of every ant is the negative direction of trend, from ant Initial position runs to the root node for studying and judging range, and the path walked leaves pheromones, i.e. correspondence in Pheromone Matrix Element+1；

Walking is completed with all ants in step 303), as the basis for estimation of end condition, is equipped with k ant user, then Carry out the positive iteration of k Pheromone Matrix；

The numerical value of each element in step 304) statistical information prime matrix, maximum numerical value is fault point position；If in result The corresponding element value in multiple paths is equal, then judges path downstream for abort situation.