Fault diagnosis method for cell group-organized P system of transmission network considering weather factors

Abstract

The present invention discloses a transmission network cell group organization type P system fault diagnosis method taking weather factors into account, which relates to the field of transmission networks. The method is based on the modeling of the power grid partition where the power outage area is located. When multiple faults or cascading faults occur, there is no need to traverse all suspected fault components. The fault diagnosis complexity does not increase with the increase of suspected fault components, which effectively reduces the complexity of the diagnosis method and has strong applicability for handling multiple faults and cascading faults. In the present cpTPS fault diagnosis model, the network topology and the protection device are coupled. Therefore, when the network topology structure changes, the model can adaptively adjust to follow the network topology change. Without changing the inference rules, correct inference after the network topology change can be achieved, and the versatility is good and convenient. The present invention is based on the increased possibility of distortion of alarm signals during transmission under disaster weather conditions. The present method takes weather factors into account in the fault diagnosis process, thereby improving the accuracy of the diagnosis method.

Description

Technical Field

[0001] The present invention relates to the field of power transmission networks, and in particular to a power transmission network cell group organization type P system fault diagnosis method taking weather factors into account. Background Art

[0002] As a crucial component of the power system, the transmission network is responsible for transmitting and distributing electrical energy. Due to its vast geographical reach and prolonged exposure to the atmosphere, transmission networks are prone to multiple and cascading failures during weather events, leading to widespread power outages. After a fault occurs, efficient and reliable fault diagnosis is paramount for rapidly restoring power supply. Therefore, accurate and rapid transmission network fault diagnosis is crucial for rapidly restoring power after a fault and maintaining the safe and stable operation of the power grid system.

[0003] Transmission network fault diagnosis involves identifying faulty components, fault type, and fault recovery. It also involves evaluating the operating behavior of protective devices and verifying the accuracy of alarm signals. To date, various transmission network fault diagnosis methods have been proposed, including expert systems, artificial neural networks, Petri nets, causal networks, Bayesian networks, and P systems. Each method has its own unique advantages and applicable application scenarios.

[0004] However, with the gradual expansion of the scale of the transmission network, the increasing complexity of its structure and the frequent occurrence of disaster weather, the problems of existing fault diagnosis methods have gradually become prominent, mainly including: (1) Existing diagnosis methods are mainly based on single-component modeling. When large-scale complex faults such as multiple faults and cascading faults occur, the number of suspected fault components increases, and fault diagnosis of multiple suspected fault components is required, so the difficulty and complexity of diagnosis increases; (2) During the passage of disaster weather, accidents such as transmission line disconnection and transmission tower collapse will cause the grid topology to change continuously, and the failure probability of each component and the system status will also change with external influences. Most of the existing diagnosis methods need to be re-modeled when the grid topology and protection configuration change, and the modeling timeliness and topology adaptability are poor; (3) Under the interference of disaster weather factors, the alarm signal is easily distorted during the communication process, thereby affecting the accuracy of the diagnosis results.

[0005] Therefore, research is urgently needed on how to efficiently diagnose multiple faults and cascading faults that are prone to occur in disaster weather, and effectively deal with problems such as false alarms and missed alarms of fault alarm signals. Summary of the Invention

[0006] In response to the above-mentioned deficiencies in the prior art, the present invention provides a transmission network cell group organization type P system fault diagnosis method taking weather factors into account, which solves the problem that the alarm signal is easily distorted during the communication process under the interference of disaster weather factors, thereby affecting the accuracy of the diagnosis results.

[0007] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0008] A method for diagnosing a P system fault in a cell group-organized transmission network taking weather factors into account is provided, comprising the following steps:

[0009] S1. Simplify the target power grid and establish a cpTPS (Tissue-like P System with Cell Populations) fault diagnosis model for each partition based on its network topology and protection device configuration information.

[0010] S2. Determine the power failure area and use the cpTPS fault diagnosis model of the partition where the power failure area is located as the target model;

[0011] S3, performing a credibility evaluation on the protection device, and correcting the initial object value of the input cell population in the target model according to the evaluation result to obtain a corrected model;

[0012] S4. Evaluate the level of disaster weather based on meteorological data;

[0013] S5. Perform topological matrix reasoning on the modified model to obtain the busbar judgment vector and the line judgment vector of the output cell group;

[0014] S6. Based on the level of the disaster weather, the bus judgment vector of the output cell group, and the line judgment vector of the output cell group, obtain the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor;

[0015] S7. Components corresponding to elements whose element values ​​are greater than or equal to the threshold in the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor are taken as faulty components to complete fault diagnosis.

[0016] The beneficial effects of the present invention are:

[0017] 1. This method is based on the grid partition modeling of the power outage area. When multiple faults or cascading faults occur, there is no need to traverse all suspected faulty components. The complexity of fault diagnosis does not increase with the increase of suspected faulty components, which effectively reduces the complexity of the diagnosis method and has strong applicability for handling multiple faults and cascading faults.

[0018] 2. In this cpTPS fault diagnosis model, the network topology is coupled with the protection device. Therefore, when the network topology changes, the model can adaptively adjust to the network topology changes. Without changing the inference rules, it can achieve correct inference after the network topology changes, which is versatile and convenient.

[0019] 3. Based on the increased possibility of distortion of alarm signals during transmission under disaster weather conditions, this method takes weather factors into account in the fault diagnosis process, thereby improving the accuracy of the diagnosis method. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 Schematic diagram of the process of this method;

[0021] Figure 2 Schematic diagram of cpTPS fault diagnosis model;

[0022] Figure 3 Schematic diagram of the credibility of the action moment;

[0023] Figure 4 Schematic diagram of the trapezoidal membership function of meteorological factors;

[0024] Figure 5 Schematic diagram of a standard IEEE39 node after partitioning in an embodiment. DETAILED DESCRIPTION

[0025] The specific embodiments of the present invention are described below to facilitate understanding of the present invention by those skilled in the art. However, it should be clear that the present invention is not limited to the scope of the specific embodiments. For those skilled in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the appended claims, these changes are obvious, and all inventions and creations utilizing the concepts of the present invention are protected.

[0026] like Figure 1 As shown, the transmission grid cell group organization type P system fault diagnosis method taking weather factors into account includes the following steps:

[0027] S1. Simplify the target power grid and establish a cpTPS fault diagnosis model for each partition based on its network topology and protection device configuration information.

[0028] S2. Determine the power failure area and use the cpTPS fault diagnosis model of the partition where the power failure area is located as the target model;

[0029] S3, performing a credibility evaluation on the protection device, and correcting the initial object value of the input cell population in the target model according to the evaluation result to obtain a corrected model;

[0030] S4. Evaluate the level of disaster weather based on meteorological data;

[0031] S5. Perform topological matrix reasoning on the modified model to obtain the busbar judgment vector and the line judgment vector of the output cell group;

[0032] S6. Based on the level of the disaster weather, the bus judgment vector of the output cell group, and the line judgment vector of the output cell group, obtain the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor;

[0033] S7. Components corresponding to elements whose element values ​​are greater than or equal to the threshold in the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor are taken as faulty components to complete fault diagnosis.

[0034] The specific method of step S1 includes the following sub-steps:

[0035] S1-1. Simplify the target power grid and partition the grid according to the weighted network partitioning principle to obtain various partitions;

[0036] S1-2. Establish a cpTPS fault diagnosis model for each partition based on the network topology and protection device configuration information of each partition. The expression of the cpTPS fault diagnosis model is:

[0037]

[0038] Where ∏ represents the cpTPS fault diagnosis model; O represents the initial object set; θ represents the cell group set, θ=[θ1,...,θ k ,...,θ n ], n is the total number of cell groups, the kth cell group θ k =(w in ,R1,R2,δ k / 1 ,...,δ k / j ,...,δ k / m ,scn);w in represents the initial object in O; R1 represents the transport rule of the cell group, which is in the form of (e,x / λ,f), where x and λ represent the objects in cell group e and cell group f respectively, and λ is an empty string; R2 represents the dissolution rule of the cell group, which is in the form of [a] e [b] f →[c] l , where a, b, and c represent the dissolution of object a in cell population e and object b in cell population f into object c in cell population l; δ k / j represents the jth cell in the kth cell group, and the cell modulation rule in the cpTPS fault diagnosis model is in represents the object value of the jth cell in the kth cell population at time T, β T It is an auxiliary object for the confidence of the action moment. represents the object value of the jth cell in the kth cell group after modulation at time T; m is the total number of cells in the kth cell group, one cell corresponds to one bus, and each bus is equipped with at least one protection device. When bus i and bus j are adjacent, the connection matrix c between bus i and bus j is ij = 1, when busbar i and busbar j are not adjacent, the connection matrix c between busbar i and busbar j ij =0,c ij ∈C; scn represents the anchoring connection relationship of cells in the k-th cell group. Cells form a strong and orderly cell group through anchoring connections. represents the meteorological survival microenvironment of the output cell population, τ is a real number on [0,1], representing the weather factor risk value of the transmission line; is a constant. When the dispatch center receives only one of the protection relay action signal and the circuit breaker action signal, otherwise Indicates the output of the permeation rule of the cell group tissue fluid. The triggering condition of the permeation rule is H * ≥H3, H * is the risk level corresponding to the current disaster weather, H3 is the third level risk; syn represents the connection status between cell groups, and the interconnected cell groups can exchange information; i in represents the input cell population; i out Represents the output cell population set;

[0039] like Figure 2 As shown in the figure, the cpTPS fault diagnosis model includes an input layer, an information fusion middle layer and an output layer connected in sequence; the input layer includes cell groups, each cell group contains m cells; the input layer and the information fusion middle layer are connected through a connection channel, and the feedback objects on the connection channel include and and Based on expert experience, the values ​​are set to 0.6 and 0.4 respectively, which are used to control the ratio of input and output of objects in the cell group. Only the multiset of objects that meet the conditions will be transmitted to the corresponding cell group through the connection channel;

[0040] The information fusion middle layer is used to obtain the bus judgment vector of the output cell group and the line judgment vector of the output cell group, and integrate the weather factors;

[0041] The output layer is used to obtain the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor. In the information fusion middle layer, since the circuit breaker controlled by the remote backup protection is not directly connected to the corresponding protected equipment, W Bs ,WLs It needs to be converted into In the output layer, It is the output cell group that takes weather factors into consideration (the line judgment vector that integrates weather factors).

[0042] The specific method of step S2 includes the following sub-steps:

[0043] S2-1, set the current search iteration number to 1;

[0044] S2-2. Construct a component number set N k : From the time the fault occurs to the time the protection relay trips and triggers the circuit breaker to clear the fault, the dispatch center will receive the circuit breaker action alarm signal, number all the circuit breakers corresponding to the circuit breaker action alarm signal and form a set N k ;

[0045] S2-3, visit N in order of number k The circuit breaker is opened and it is determined whether the current circuit breaker has been searched. If so, the next circuit breaker is searched along the power-off side of the circuit breaker and the number of search iterations is increased by 1; otherwise, the process goes to step S2-4;

[0046] S2-4. Determine whether the current circuit breaker is a fault area boundary circuit breaker. If so, put the corresponding number of the circuit breaker into the component number set η. k If yes, go to step S2-5; otherwise, return to step S2-3 and increase the number of search iterations by 1;

[0047] S2-5, determine whether the search is complete N k If all circuit breakers are found, go to step S2-6; otherwise, return to step S2-3 and increase the number of search iterations by 1;

[0048] S2-6. Visit η in order of number k , along η k Search the circuit breaker in the power outage direction and record it. If the circuit breaker found is a circuit breaker at the fault area boundary, add the number corresponding to the circuit breaker to η k ;

[0049] S2-7. Get current η k All passive networks where the circuit breaker is located are included, and the obtained passive network is used as the power-off area, and the cpTPS fault diagnosis model of the partition where the power-off area is located is used as the target model.

[0050] The specific method of step S3 includes the following sub-steps:

[0051] S3-1. Collect the action information of the protection device in the SCADA system and obtain the fault time β through the alarm signal time sequence reasoning. time ;

[0052] S3-2, extract the characteristics of the voltage and current waveforms within the set distance of the protection device at the time of action, and obtain the fault moment β through the changes in the voltage and current waveforms. elec ;

[0053] S3-3, β time and β elec Perform comparisons and obtain the true state of the protection device action based on the comparison results; construct and obtain the credibility of the protection device action state based on the protection device relay confidence table and the corresponding circuit breaker confidence table;

[0054] S3-4. Obtaining the reliability of the protection device's action time based on the protection device's action delay;

[0055] S3-5, according to the formula:

[0056]

[0057] Get the object value of the i-th cell in the k-th cell group at the initial moment That is, the corrected initial object value of the i-th cell in the k-th cell population; and Both are weight coefficients and can be set to 0.55 and 0.45 respectively; is the credibility of the action state of the protection device corresponding to the i-th cell in the k-th cell group; time (i) is the reliability of the action time of the protection device corresponding to the i-th cell in the k-th cell group.

[0058] In step S3-3, β time and β elec The specific method for comparing and obtaining the actual state of the protection device action based on the comparison results is as follows:

[0059] If β time and β elec Inconsistency means that the voltage and current waveforms within the set distance of the protection device have not changed, but the dispatching center receives the relay action signal, which means that the two information do not match, indicating that the protection relay has malfunctioned, but the circuit breaker has not operated;

[0060] If β time and β elec If they are consistent, the action is judged to be correct;

[0061] If β time and β elecInconsistency, the voltage and current waveforms within the set distance of the protection device change, and β elec If the dispatch center does not receive the relay action signal at the same time as the circuit breaker actuation, it is determined that the relay has not actuated, but the circuit breaker has malfunctioned, and the circuit breaker has cut off the circuit, causing the voltage and current waveforms near the protection device to change, rather than a fault causing the voltage and current waveforms near the protection device to change;

[0062] If β time and β elec Inconsistency, the voltage and current waveforms within the set distance of the protection device change, and β elec When the dispatch center receives the relay action signal at the same time as the circuit breaker operation, it determines that the relay has malfunctioned and drives the circuit breaker to operate, cutting off the circuit and causing the voltage and current waveforms near the protection device to change, rather than causing the voltage and current waveforms near the protection device to change due to a fault.

[0063] The protection device relay confidence table and the corresponding circuit breaker confidence table are shown in Table 1.

[0064] Table 1

[0065]

[0066] The credibility of the action state of the protection device can be obtained according to the actual action state of the relay and circuit breaker and the credibility of the relay and circuit breaker respectively.

[0067] The specific method of step S3-4 is: according to the formula:

[0068]

[0069] Obtain the reliability of the protection device action time time ; Where t represents the time when the relay or circuit breaker operates; t e Indicates the time when the e-th relay or circuit breaker operates; Δt e The time parameters involved in the reliability of the protection device action time are as follows: Figure 3 Time delay refers to the time lag between a protection device receiving a signal and mechanically actuating, and different protection devices have different delay times. Table 2 shows the time delays for various protection devices.

[0070] Table 2

[0071] Corresponding protective device Action delay / ms Line main protection Δt=[10,20] Busbar main protection Δt=[10,20] breaker Δt=[40,60] Line close backup protection Δt=[190,240] Line remote backup protection Δt=[1950,2050]

[0072] The specific method of step S4 includes the following sub-steps:

[0073] S4-1. Construct a failure rate table for meteorological factors at different environmental levels. Obtain real-time meteorological data for the fault area from a weather station. Combined with the failure rate table, determine the failure rate for the meteorological factor at the time of the fault. The failure rate table for meteorological factors at different environmental levels is shown in Table 3. The meteorological factors used in this implementation include strong winds, ice cover, lightning, rain, and snow.

[0074] Table 3

[0075]

[0076] S4-2, according to the formula:

[0077]

[0078] Get the fuzzy judgment value ν under the ath meteorological factor and the bth risk level ab , and then get the fuzzy evaluation matrix Z,ν ab ∈Z; where p a represents the failure rate corresponding to the ath meteorological factor. The failure rates corresponding to strong wind, icing, lightning, rain, and snow are shown in Table 4. y1 represents the boundary value of the evaluation standard for the first-level risk H1; y2 and y3 represent the boundary values ​​of the evaluation standard for the second-level risk H2; y4 and y5 represent the boundary values ​​of the evaluation standard for the third-level risk H3; y6 and y7 represent the boundary values ​​of the evaluation standard for the fourth-level risk H4; y8 represents the boundary value of the evaluation standard for the fifth-level risk H5.

[0079] Table 4

[0080]

[0081] S4-3. Obtain the pairwise relationship matrix A, A between meteorological factors ac ∈A,A ac Indicates the relative importance of the a-th meteorological event compared to the c-th meteorological event. Its value (importance scale) is shown in Table 5.

[0082] Table 5

[0083]

[0084] S4-4, according to the formula:

[0085]

[0086] Get the weight value u of the a-th weather a ; Where N represents the total number of meteorological categories;

[0087] S4-5, according to the formula:

[0088] H=Z·U

[0089] Obtain the fuzzy comprehensive evaluation matrix H, and obtain the risk level corresponding to the current disaster weather according to the maximum membership principle based on the fuzzy comprehensive evaluation matrix H, and obtain the corresponding weather factor risk value τ according to the risk level corresponding to the current disaster weather. The weather factor risk values ​​corresponding to the risk levels of different disaster weather are shown in Table 6. The trapezoidal membership function of the meteorological factors in this embodiment is as follows: Figure 4 shown.

[0090] Table 6

[0091]

[0092] The specific method of step S5 includes the following sub-steps:

[0093] S5-1. Establish a connection matrix C based on the system bus topology, and construct the bus main protection connection matrix PR based on the connection matrix C. Bm , Line main protection connection matrix PR Lm , Line near backup protection PR Lp , line remote backup protection PR Ls , and PR Bm The corresponding circuit breaker connection matrix CB Bm , and PR Lm The corresponding circuit breaker connection matrix CB Lm , and PR Lp The corresponding circuit breaker connection matrix CB Lp , and PR Ls The corresponding circuit breaker connection matrix CB Ls ; If the cell population θ k (1≤k≤n) intracellular δ k / i The corresponding busbar and cell δ k / j If there is a connection relationship between the corresponding buses, Cell δ k / i The busbar main protection, line main protection, line near backup protection, and line far backup protection object values ​​on the side; otherwise If the cell population θ k (1≤k≤n) intracellular δ k / i The corresponding busbar and cell δ k / j If there is a connection relationship between the corresponding buses, Cell δ k / i The busbar main protection on the side corresponds to the circuit breaker, the line main protection corresponds to the circuit breaker, the line near backup protection corresponds to the circuit breaker, and the line far backup corresponds to the circuit breaker object value; otherwise

[0094] S5-2, according to the formula:

[0095]

[0096]

[0097]

[0098]

[0099] Obtain the connection matrix W of the busbar main protection and the corresponding circuit breaker after information fusion after the cell group is dissolved Bm , the connection matrix W after information fusion between the line main protection and the corresponding circuit breaker Lm , the connection matrix W after information fusion between the line backup protection and the corresponding circuit breaker Lp The connection matrix W after information fusion between the remote backup protection and the corresponding circuit breaker Ls ; If the cell population θ k (1≤k≤n) intracellular δ k / i The corresponding busbar and cell δ k / j If there is a connection relationship between the corresponding buses, Cell δ k / i The busbar main protection, line main protection, line near backup protection, line far backup protection and the object value of the corresponding circuit breaker after cell group dissolution; otherwise in in

[0100] S5-3. Judgment Is it true? If so, according to the formula Get the converted line remote backup protection connection matrix The value of the object in row j, column i Otherwise, according to the formula Get the converted line remote backup protection connection matrix The value of the object in row j, column i Where W Ls (j,i) represents W Ls The object value at row j and column i; W Ls (r,i) represents W Ls The object value in row r and column i; c ij Represents the object value of the i-th row and j-th column of the connection matrix C; Indicates the multiplication of the elements at each corresponding position in the two matrices; c gi Represents the object value of the g-th row and i-th column of the connection matrix C;

[0101] S5-4, according to the formula:

[0102]

[0103] Get the converted busbar remote backup protection connection matrix in(·) T Represents the transpose of a matrix;

[0104] If the cell population θ k (1≤k≤n) intracellular δ k / i The corresponding busbar and cell δ k / j If there is a connection relationship between the corresponding buses, They are The object value after special mapping conversion rules; otherwise

[0105] S5-5, according to the formula:

[0106]

[0107]

[0108] Get the busbar judgment vector W of the output cell group respectively B And the circuit judgment vector W of the output cell group L ;in Indicates that the maximum value of each element at each corresponding position in the two matrices is obtained; E m×1 represents an m-dimensional unit column vector; Represents an extraction function that extracts the non-zero elements in the upper right triangle of a matrix from left to right. is a real number on (0,1). in

[0109] The specific method of step S6 is: determine whether the current disaster weather level reaches the preset level (the preset level is preferably the third level risk H3), if so, according to the formula and Get the busbar judgment vector of the output cell group fusion weather factor respectively And the line judgment vector of the output cell group fusion weather factor Otherwise, just W B As the busbar judgment vector of the output cell group fusion weather factor Directly W L As the line judgment vector of the output cell group fusion weather factor in is an m-dimensional column vector. When the dispatch center receives only one of the protection relay action signal and the circuit breaker action signal, The value of the corresponding element in is 0.4; otherwise the value of the corresponding element is 0; that is

[0110] In a specific implementation process, the threshold in step S7 is 0.6.

[0111] In one embodiment of the present invention, a standard IEEE 39 node system is used as a diagnosis object, and Example 1 is taken as an example to provide a specific calculation process to facilitate detailed understanding. Figure 5 The network topology of the standard IEEE39 node after partitioning is shown in Table 7. The preset fault scenarios and faulty component results of Example 1 are shown in Table 7.

[0112] Table 7

[0113]

[0114] First, after the fault occurs, the breadth-first search method is used to determine that the power failure area is located in zone S1. Figure 5 The connection matrix C of the S1 partition is:

[0115]

[0116] In order to ensure the accuracy and authenticity of fault diagnosis, the operating status and confidence of the protection device are evaluated respectively.

[0117] Table 8

[0118]

[0119] From Table 8, we can see that the circuit breaker action time is consistent with the voltage and current waveform change time, and no protection relay action signal is received. The alarm signal timing reasoning obtains the fault time β time Less than the fault moment β obtained by observing the changes in voltage and current waveforms elec , thus inferring that circuit breaker CB60 is malfunctioning. Correct the initial object value of the input cell group.

[0120] Based on the connection matrix C and the confidence evaluation results of the protection device, the busbar main protection connection matrix PR is established. Bm , line main protection connection matrix PR Lm , line near backup protection connection matrix PR Lp , line remote backup protection connection matrix PR Ls And the corresponding circuit breaker connection matrix CB Bm ,CB Lm ,CB Lp ,CBLs :

[0121]

[0122]

[0123]

[0124]

[0125]

[0126]

[0127]

[0128]

[0129] Combining the real-time meteorological information with Table 3, we can get the failure rate of each meteorological factor in this area as (0.44, 0.0032, 0.52, 0.08, 0.003). The fuzzy evaluation matrix of meteorological factors is obtained as follows: The pairwise relationship matrix A between meteorological factors is: Solving the matrix, the weight vector of the meteorological factors is U = [0.2591 0.0566 0.4637 0.1811 0.0395]; the fuzzy comprehensive evaluation matrix of the meteorological factors is H = [0.0955 0.0006 0 0.1252 0.7787]; based on Table 6 and the maximum membership principle, the meteorological factor risk level is H5, and the weather factor risk value is 1.

[0130] The corresponding They are:

[0131]

[0132]

[0133] The output object values ​​of line L28 and bus B23 are 0.9156 and 0.9149 respectively, indicating that line L28 and bus B23 are faulty.

[0134] In addition, to test the reliability of the diagnostic performance of the cpTPS fault diagnosis model, a fault information uncertainty experiment was conducted based on the IEEE39 node. The probability of the occurrence of uncertain information κ was increased in the case of single fault, double fault, triple fault, quadruple fault, and quintuple fault. 100,000 random tests were performed, and the accuracy of the method was statistically calculated as shown in Table 9.

[0135] Table 9

[0136]

[0137] Where L represents a single line fault, B represents a single busbar fault, and 1L1B represents a dual fault involving a line and a busbar. The other parameters can be derived similarly. Table 9 shows that when the alarm signal is uncertain (i.e., false alarms or missed alarms occur), even when a quadruple or quintuple fault occurs with an uncertainty probability of κ of 14%, this method still achieves high accuracy.

[0138] In summary, the present invention is based on the modeling of the power grid partition where the power outage area is located. When multiple faults or cascading faults occur, there is no need to traverse all suspected fault components, and the complexity of fault diagnosis will not increase with the increase of suspected fault components, which effectively reduces the complexity of the diagnosis method and has strong applicability for handling multiple faults and cascading faults; in this cpTPS fault diagnosis model, the network topology and the protection device are coupled, so when the network topology structure changes, the model can adaptively adjust to follow the network topology changes, and correct reasoning after the network topology changes can be achieved without changing the reasoning rules, which has good versatility and convenience; the present invention is based on the increased possibility of distortion of alarm signals during transmission under disaster weather conditions. This method takes weather factors into consideration in the fault diagnosis process, thereby improving the accuracy of the diagnosis method.

Claims

1. A method for diagnosing faults in a cell group-organized P system of a transmission network taking weather factors into account, characterized in that: The following steps are involved: S1. Simplify the target power grid and establish a cell cluster organization type P system fault diagnosis model for each partition based on its network topology and protection device configuration information; S2. Determine the power-off area and use the cell group organization type P system fault diagnosis model of the partition where the power-off area is located as the target model; S3, performing a credibility evaluation on the protection device, and correcting the initial object value of the input cell population in the target model according to the evaluation result to obtain a corrected model; S4. Evaluate the level of disaster weather based on meteorological data; S5. Perform topological matrix reasoning on the modified model to obtain the busbar judgment vector and the line judgment vector of the output cell group; S6. Based on the level of the disaster weather, the bus judgment vector of the output cell group, and the line judgment vector of the output cell group, obtain the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor; S7. Components corresponding to elements whose element values ​​are greater than or equal to the threshold in the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor are taken as faulty components to complete fault diagnosis.

2. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 1, characterized in that: The specific method of step S1 includes the following sub-steps: S1-1. Simplify the target power grid and partition the grid according to the weighted network partitioning principle to obtain various partitions; S1-2. Establish a cell cluster organization type P system fault diagnosis model for each partition based on the network topology and protection device configuration information of each partition. The expression of the cell cluster organization type P system fault diagnosis model is: in It represents the fault diagnosis model of cell group organization type P system; Represents the initial object set; Represents a collection of cell populations, , n is the total number of cell populations, k cell populations ; express The initial object in Represents the transport rules of cell populations; Indicates the rules for the dissolution of a cell population; Indicates the k In the cell population j cells; m For the k The total number of cells in a cell group, one cell corresponds to one busbar, and each busbar is equipped with at least one protection device. i and busbar j When adjacent, busbar i and busbar j The connection matrix , when the bus i and busbar j When not adjacent, the busbar i and busbar j The connection matrix , ; Indicates the k Anchorage connection relationship of cells within a cell group; represents the meteorological survival microenvironment of the output cell population, , is a real number on [0,1], representing the weather factor risk value of the transmission line; is a constant. When the dispatch center receives only one of the protection relay action signal and the circuit breaker action signal, ,otherwise ; Indicates the osmotic rules of the tissue fluid of the output cell group; Indicates the connectivity status between cell populations; represents the input cell population set; Represents the output cell population set; The cell group organization type P system fault diagnosis model includes an input layer, an information fusion intermediate layer and an output layer connected in sequence; the input layer includes cell groups, each cell group contains m cells; The input layer and the information fusion middle layer are connected through a connection channel. The feedback objects on the connection channel include and , and are all constants used to control the ratio of input and output of objects in the cell group; The information fusion middle layer is used to obtain the bus judgment vector of the output cell group and the line judgment vector of the output cell group, and integrate the weather factors; The output layer is used to obtain the bus judgment vector of the output cell group fusion weather factor and the line judgment vector of the output cell group fusion weather factor.

3. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 1, characterized in that: The specific method of step S2 includes the following sub-steps: S2-1, set the current search iteration number to 1; S2-2. Construct a component number set :From the time the fault occurs to the time the protection relay trips and triggers the circuit breaker to clear the fault, the dispatch center will receive the circuit breaker action alarm signal, number all the circuit breakers corresponding to the circuit breaker action alarm signal and form a collection ; S2-3, visit in order of number The circuit breaker is opened and it is determined whether the current circuit breaker has been searched. If so, the next circuit breaker is searched along the power-off side of the circuit breaker and the number of search iterations is increased by 1; otherwise, the process goes to step S2-4; S2-4. Determine whether the current circuit breaker is a fault area boundary circuit breaker. If so, put the corresponding number of the circuit breaker into the component number set. If yes, go to step S2-5; otherwise, return to step S2-3 and increase the number of search iterations by 1; S2-5, determine whether the search is complete If all circuit breakers are found, go to step S2-6; otherwise, return to step S2-3 and increase the number of search iterations by 1; S2-6, visit in order of number ,along Search the circuit breaker in the power outage direction and record it. If the circuit breaker found is the fault area boundary circuit breaker, add the number corresponding to the circuit breaker to the fault area boundary circuit breaker. ; S2-7. Get the current All passive networks where the circuit breaker is located are selected, and the obtained passive network is taken as the power-off area, and the cell group organization type P system fault diagnosis model of the partition where the power-off area is located is taken as the target model.

4. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 1, characterized in that: The specific method of step S3 includes the following sub-steps: S3-1. Collect the action information of the protection device in the SCADA system and obtain the fault time through the alarm signal time sequence reasoning. ; S3-2. Extract the characteristics of the voltage and current waveforms within the set distance of the protection device at the time of action, and obtain the fault time through the changes in the voltage and current waveforms. ; S3-3, will and Perform comparison and obtain the actual status of the protection device action based on the comparison results; Construct and obtain the operating state credibility of the protection device based on the protection device relay confidence table and the corresponding circuit breaker confidence table; S3-4. Obtaining the reliability of the protection device's action time based on the protection device's action delay; S3-5, according to the formula: Get the k The first i The object value of each cell at the initial moment , that is, k The first i The initial object value after the cells are corrected; and All are weight coefficients; For the k The first i Credibility of the action status of the protection device corresponding to each cell; For the k The first i The reliability of the action time of the protection device corresponding to each cell.

5. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 4, characterized in that: In step S3-3, and The specific method for comparing and obtaining the actual state of the protection device action based on the comparison results is as follows: like and Inconsistency: the voltage and current waveforms within the set distance of the protection device have not changed, but the dispatching center receives the relay action signal, and it is determined that the relay has malfunctioned and the circuit breaker has not operated; like and If they are consistent, the action is judged to be correct; like and Inconsistency, the voltage and current waveforms within the set distance of the protection device change, and If the dispatch center does not receive the relay action signal at the same time as the circuit breaker actuation, it is determined that the relay has not actuated, but the circuit breaker has malfunctioned, and the circuit breaker has cut off the circuit, causing the voltage and current waveforms near the protection device to change, rather than a fault causing the voltage and current waveforms near the protection device to change; like and Inconsistency, the voltage and current waveforms within the set distance of the protection device change, and When the dispatch center receives the relay action signal at the same time as the circuit breaker operation, it determines that the relay has malfunctioned and drives the circuit breaker to operate, cutting off the circuit and causing the voltage and current waveforms near the protection device to change, rather than causing the voltage and current waveforms near the protection device to change due to a fault.

6. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 4, characterized in that: The protection device relay confidence table is as follows: The confidence level for the main protection of the busbar is 0.86; the confidence level for the main protection of the line is 0.99; the confidence level for the local backup protection of the line is 0.8; the confidence level for the remote backup protection of the line is 0.7; all confidence levels for unoperated protection are 0.2; The corresponding circuit breaker confidence table is as follows: the confidence level of the main protection for the busbar is 0.98; the confidence level of the main protection for the line is 0.98; the confidence level of the local backup protection for the line is 0.85; the confidence level of the remote backup protection for the line is 0.75; and all confidence levels of the unoperated are 0.

2.

7. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 4, characterized in that: The specific method of step S3-4 is: According to the formula: Obtaining the reliability of the protection device action time ;in t Indicates the moment when the relay or circuit breaker operates; Indicates the e The moment a relay or circuit breaker operates; Indicates the e The time delay allowed by a relay or circuit breaker; Indicates the allowed delay for an event.

8. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 1, characterized in that: The specific method of step S4 includes the following sub-steps: S4-1. Construct a failure rate table for meteorological factors at different environmental levels. Obtain real-time meteorological data of the fault area through a weather station. Combined with the failure rate table, obtain the failure rate of the meteorological factor corresponding to the fault time. S4-2, according to the formula: Get the a Meteorological factors, b Fuzzy judgment value under risk level , and then we get the fuzzy evaluation matrix Z, ;in Indicates the a Failure rate corresponding to each meteorological factor; Indicates level 1 risk The boundary value of the judgment standard; and Indicates secondary risk The judgment standard boundary value; and Indicates level 3 risk The boundary value of the judgment standard; and Indicates level 4 risk The boundary value of the judgment standard; Indicates level 5 risk The judgment standard boundary value; S4-3. Obtaining the pairwise relationship matrix between meteorological factors , , Indicates the a Compared with the first c The relative importance of individual weather conditions; S4-4, according to the formula: Get the a The weight of each weather ;in N Indicates the total number of meteorological categories; S4-5, according to the formula: Obtain the fuzzy comprehensive evaluation matrix H, and obtain the risk level corresponding to the current disaster weather according to the maximum membership principle based on the fuzzy comprehensive evaluation matrix H, and obtain the corresponding weather factor risk value according to the risk level corresponding to the current disaster weather .

9. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 2, characterized in that: The specific method of step S5 includes the following sub-steps: S5-1. Establish a connection matrix based on the system bus topology , according to the connection matrix Construct busbar main protection connection matrix , Line main protection connection matrix , Line near backup protection , line remote backup protection ,and Corresponding circuit breaker connection matrix ,and Corresponding circuit breaker connection matrix ,and Corresponding circuit breaker connection matrix , and with Corresponding circuit breaker connection matrix ; S5-2, according to the formula: Obtain the connection matrix of the busbar main protection and the corresponding circuit breaker after information fusion after the cell group is dissolved , the connection matrix of the line main protection and the corresponding circuit breaker after information fusion , the connection matrix of line backup protection and corresponding circuit breaker after information fusion The connection matrix after information fusion between the remote backup protection of the line and the corresponding circuit breaker ; S5-3. Judgment Is it true? If so, according to the formula Get the converted line remote backup protection connection matrix Middle OK The object value of the column Otherwise, according to the formula Get the converted line remote backup protection connection matrix Middle OK The object value of the column ;in express No. OK The object value of the column; express No. OK The object value of the column; Represents the connection matrix No. OK The object value of the column; Represents the multiplication of the elements at each corresponding position in the two matrices; Represents the connection matrix No. OK The object value of the column; S5-4, according to the formula: Get the converted busbar remote backup protection connection matrix ;in Represents the transpose of a matrix; S5-5, according to the formula: Get the busbar judgment vector of the output cell group respectively And the circuit judgment vector of the output cell group ;in Indicates that the maximum value of each element at each corresponding position in the two matrices is obtained; express m -dimensional unit column vector; Represents an extraction function that extracts the non-zero elements in the upper right triangle of a matrix from left to right.

10. The method for diagnosing faults of a cell group-organized P system of a power transmission network taking weather factors into account according to claim 9, characterized in that: The specific method of step S6 is: Determine whether the current disaster weather level reaches the preset level. If so, according to the formula and Get the busbar judgment vector of the output cell group fusion weather factor respectively And the line judgment vector of the output cell group fusion weather factor Otherwise, just As the busbar judgment vector of the output cell group fusion weather factor , directly As the line judgment vector of the output cell group fusion weather factor ; in is the weather factor risk value corresponding to the current disaster weather level; for m dimensional column vector, when the dispatch center receives only one of the protection relay action signal and the circuit breaker action signal, The value of the corresponding element in is 0.4; otherwise the value of the corresponding element is 0; that is .

Images