Complex power grid fault time-lag influence intensity calculation method, device and equipment

Abstract

The invention provides a method, a device and equipment for calculating time-lag influence intensity of a complex power grid fault, and relates to the technical field of power systems. The method comprises the following steps: determining a first statistical magnitude corresponding to a first device, a second device and a preset device interval number and a second statistical magnitude corresponding to the first device and the preset device interval number in a plurality of device power failure event sequences; determining the ratio of the first statistical magnitude to the second statistical magnitude as a first estimated value corresponding to the interval number of the first equipment, the second equipment and the preset equipment; and iteratively updating the first estimation value by using a gradient ascending method and a plurality of equipment power failure event sequences to obtain a second estimation value corresponding to the first equipment, the second equipment and the preset equipment interval number, and taking the second estimation value as the time-delay fault influence intensity of the first equipment on the second equipment under the preset equipment interval number. According to the invention, the accuracy of the second estimated value can be effectively improved, so that the accuracy of the calculated time-delay fault influence intensity is improved.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method, apparatus and equipment for calculating the intensity of the time-delay effect of complex power grid faults. Background Technology

[0002] As a complex interconnected system supporting social production and daily life, the power grid's power outage risks are often not caused by a single factor, but rather by a chain reaction of interconnected equipment outage events. These equipment failures may begin with localized line tripping or transformer failures, gradually spreading to surrounding areas and ultimately threatening the stable operation of the entire power grid. To effectively mitigate these outage risks, it is necessary to utilize professional monitoring and analysis methods to accurately predict the development process of multiple equipment outage events, clarify the fault propagation path and scope of impact, and thus promptly implement targeted corrective measures such as load transfer and backup equipment activation to minimize the economic and social losses caused by power outages.

[0003] Currently, there are two main approaches to studying the development process of complex power grid faults: physical simulation models and statistical probability models. Physical simulation models primarily simulate the physical process of faults by solving power flow equations or differential-algebraic equations. However, these models are computationally intensive and struggle to efficiently generate a large number of fault events for analysis, thus limiting their application. Statistical probability models, on the other hand, use statistical parameters such as fault impact probability matrices to predict fault development trends, and are widely studied due to their lower computational load.

[0004] However, when statistical probability models are used to study the development process of complex power grid faults, they rely on the Markov assumption and only consider the impact of the previous fault on a single equipment outage event, while ignoring the potential impact of equipment outage events that occurred earlier. This simplification makes it impossible for the model to fully trace the root cause and development of time-delayed faults when analyzing complex faults, and the final output of the impact intensity judgment results deviates significantly from the actual situation. Summary of the Invention

[0005] This invention provides a method, apparatus, and equipment for calculating the impact intensity of time-delayed faults in complex power grids, in order to solve the problem that statistical probability models only consider the impact of the previous fault and ignore the impact of earlier faults, resulting in inaccurate calculated impact intensity.

[0006] In a first aspect, embodiments of the present invention provide a method for calculating the intensity of the time-delay effect of complex power grid faults, including: In a sequence of multiple equipment outage events, determine the first statistic corresponding to the first equipment, the second equipment, and the preset equipment interval number, and the second statistic corresponding to the first equipment and the preset equipment interval number; wherein, the preset equipment interval number is the number of equipment in the power grid that fails after the first equipment fails, and the second equipment is the equipment in the power grid that fails at the preset equipment interval number after the first equipment fails; The ratio of the first statistic to the second statistic is determined as the first estimated value corresponding to the number of intervals between the first device, the second device, and the preset device. Using the gradient ascent method and multiple equipment outage event sequences, the first estimate is iteratively updated to obtain the second estimate corresponding to the first equipment, the second equipment, and the preset equipment interval number. This second estimate is then used as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset equipment interval number.

[0007] Secondly, embodiments of the present invention provide a calculation device for the intensity of the time-delay effect of complex power grid faults, comprising: The determination module is used to determine the first statistic corresponding to the first device, the second device, and the preset device interval number in a sequence of multiple device power outage events, and the second statistic corresponding to the first device and the preset device interval number; wherein, the preset device interval number is the number of devices in the power grid that fail after the first device fails, and the second device is the device in the power grid that fails at the preset device interval number after the first device fails; The calculation module is used to determine the ratio of the first statistic and the second statistic as a first estimated value corresponding to the number of intervals between the first device, the second device, and the preset device; The update module is used to iteratively update the first estimate using the gradient ascent method and multiple equipment outage event sequences to obtain the second estimate corresponding to the first equipment, the second equipment, and the preset equipment interval number, and to use it as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset equipment interval number.

[0008] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the method described in the first aspect or any possible implementation thereof.

[0009] In this embodiment of the invention, a first statistic corresponding to a first device, a second device, and a preset number of device intervals is determined through a sequence of multiple power outage events in the power grid. A second statistic corresponding to the first device and the preset number of device intervals is also determined. A first estimated value is determined based on the first and second statistics, taking into account the time-delayed fault impact of a first device failure on a second device failure occurring after a preset number of device intervals. The first estimated value is iteratively updated using the gradient ascent method to obtain a second estimated value, which takes into account the impact of other device failures during the development of the second device failure after the first device failure has passed the preset number of device intervals. This effectively improves the accuracy of the second estimated value, thereby improving the accuracy of the time-delayed fault impact intensity. This provides power grid dispatchers with a comprehensive and reliable fault risk reference, thereby effectively enhancing the resilience and safe operation capability of the power grid against faults. Attached Figure Description

[0010] Figure 1 This is a flowchart illustrating the implementation of the method for calculating the impact intensity of time delay in complex power grid faults provided in this embodiment of the invention. Figure 2 This is a flowchart illustrating the implementation of step S130 of the method for calculating the impact intensity of time delay in complex power grid faults provided in this embodiment of the invention. Figure 3 This is a schematic diagram of the structure of the calculation device for the influence intensity of time delay of complex power grid faults provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0011] It should be noted that, since the analysis aims to determine the cascading effects of a fault in a power grid device on other devices after a certain time delay, the analysis must consider the faulty device, the number of devices in the intermediate intervals, and the devices affected by the faulty device. This cascading effect can be described by constructing a time-delay probability parameter tensor. Because considering the impact requires information from three dimensions, the time-delay probability parameter tensor is a three-dimensional tensor. T ∈R N×N×kmax ,in: N This represents the number of power outage events affecting equipment in the power grid. k max The preset maximum time delay length, such as k max =3, which corresponds to a maximum of 3 events that a single equipment outage event can affect after the event occurs.

[0012] In some embodiments, a three-dimensional tensor includes multiple tensor elements, and tensor elements T i,j,kThis indicates that the coordinate indices of the 1st, 2nd, and 3rd dimensions in the tensor are respectively taken as... i , j , k The element value at time, its meaning is when the device i A malfunction caused the equipment i The direct shutdown also indirectly affected... k - If one of the other devices malfunctions, causing other devices to indirectly stop operating, the chain reaction will affect the next... k The malfunctioning device was precisely the equipment... j The conditional probability is expressed as the device i Power outages caused by faults affect equipment j The time delay of the fault and the intensity of its impact. i , j , k It is a positive integer, which can be set according to the number of devices in the power grid. Different tensor elements in the three-dimensional tensor can realize the impact of instantaneous faults ( k =1) to the impact of time-delay faults ( k >1) Characterization of the impact of failures throughout the entire process. The solution in this application can calculate the value of each tensor element in the three-dimensional tensor.

[0013] The embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0014] See Figure 1 The document illustrates a flowchart of the implementation of a method for calculating the impact intensity of time-delayed faults in complex power grids, as provided in an embodiment of the present invention. Details are as follows: Step S110: Determine the first statistic corresponding to the first device, the second device, and the preset device interval number in the multiple device outage event sequences, and the second statistic corresponding to the first device and the preset device interval number; wherein, the preset device interval number is the number of devices in the power grid that fail after the first device fails, and the second device is the device in the power grid that fails at the preset device interval number after the first device fails.

[0015] In some embodiments, the preset device interval number is a positive integer. The device outage event sequence includes multiple faulty devices that experienced outage events and the order in which the outage events of each faulty device occurred, that is, the fault sequence of each faulty device. The device outage event sequence is extracted from historical power data of the power grid.

[0016] It should be noted that the first device can be any device in the power grid, and the second device can also be any device in the power grid. The first statistic refers to the total number of times the event "after the first device fails, a device at a preset interval fails, and the last device to fail is exactly the second device" occurs in all device outage event sequences. The second statistic refers to the total number of times the event "after the first device fails, a device at a preset interval fails" occurs in all device outage event sequences.

[0017] In one possible implementation, step S110 is specifically processed as follows: For each equipment power outage event sequence, if the sequence contains fault information of a first device, and a second device fails after a preset equipment interval following the first device failure, then the sequence is identified as a first target sequence; the number of first target sequences is identified as a first statistic. For each equipment power outage event sequence, if the sequence contains fault information of a first device, and a power grid failure occurs after a preset equipment interval following the first device failure, then the sequence is identified as a second target sequence; the number of second target sequences is identified as a second statistic.

[0018] In some embodiments, when calculating the first statistic, it is necessary to determine whether each device power outage event sequence contains the event "after the first device fails, a device with a preset device interval fails, and the last device to fail happens to be the second device", and the device power outage event sequence containing this event is determined as the first target sequence, and the total amount of the first target sequence is determined as the first statistic.

[0019] When calculating the second statistic, it is necessary to determine whether each equipment power outage event sequence contains the event "after the first equipment fails, a set number of equipment fails at a predetermined interval", and the equipment power outage event sequence containing this event is determined as the second target sequence, and the total amount of the second target sequence is determined as the second statistic.

[0020] Step S120: The ratio of the first statistic and the second statistic is determined as the first estimated value corresponding to the number of intervals between the first device, the second device, and the preset device.

[0021] In some embodiments, the formula for calculating the first estimate is:

[0022] in, This is the first estimate. As the first statistic, This is the second statistic. As the first piece of equipment, For the second device, This is the preset number of equipment intervals.

[0023] Step S130: Using the gradient ascent method and multiple equipment outage event sequences, the first estimated value is iteratively updated to obtain the second estimated value corresponding to the first equipment, the second equipment, and the preset equipment interval number, which is used as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset equipment interval number.

[0024] In some embodiments, the gradient ascent method is combined with multiple equipment outage event sequences to iteratively update the first estimate, obtaining the time-delayed fault impact intensity of the first device on the second device under a preset number of device intervals. This allows for the simultaneous consideration of the time-delayed impact of multiple devices when determining the time-delayed fault impact intensity, thus improving the accuracy of the time-delayed fault impact intensity. The time-delayed fault impact intensity is a quantitative indicator used to measure the impact of a fault in a power grid on the normal operation of subsequent devices as the fault propagates over time. The time-delayed fault impact intensity is not only related to the severity of the fault in the first device itself, but also depends on the time interval between the fault impact propagation from the first device to subsequent devices, the real-time power flow distribution and load level of the power grid, and the tolerance threshold and protection configuration of subsequent devices. A higher time-delayed fault impact intensity indicates that even with a delay in the fault impact, it is easier to exceed the operating limits of subsequent devices, thereby triggering a cascading fault. Conversely, a lower time-delayed fault impact intensity means that the delayed propagation of the fault is more moderate, and the power grid has more time to initiate preventive measures to prevent the fault from spreading.

[0025] See Figure 2 The specific processing method of step S130 above includes steps S1301-S1302, the details of which are as follows: Step S1301: For each power outage event sequence, if the t-th faulty device in the sequence is the first device and the (t+k)-th faulty device is the second device, then the sequence is determined as a valid sequence; where t is a positive integer and k is a preset device interval.

[0026] In some embodiments, the values ​​of t and k can be set manually. When a power outage event sequence contains the event "the first device is the t-th device in the sequence, after the first device fails, a device with a preset interval number (k) fails, and the last device to fail is the second device", the power outage event sequence is a valid sequence.

[0027] Step S1302: Based on multiple valid sequences and the second statistic, the first estimate is iteratively updated using the gradient ascent method to obtain the second estimate.

[0028] In some embodiments, the second estimate differs from the first estimate. The first estimate is the ratio of the first statistic to the second statistic, while the second estimate is obtained by first updating the first statistic using the gradient ascent method, then calculating the ratio of the updated first statistic to the second statistic, and calculating the error between each iteration and the previous iteration, continuously updating until the final value converges. Since the operating state of the power grid differs in different sequences, and the faulty devices between the first and second devices differ, the degree of influence of the first device on the second device will vary in different sequences. Therefore, the second estimate, through iterative updates, takes into account the differences in the degree of influence of the first device on the second device in different sequences, making the second estimate more accurate than the first estimate.

[0029] In one possible implementation, step S1302 is specifically processed as follows: Before iteration, the first statistic is initialized to 0; Looping steps: based on multiple valid sequences and the first estimate, the first statistic is updated, and based on the updated first and second statistics, the first estimate is updated to obtain the third estimate; the difference between the third estimate and the first estimate is calculated; if the difference is not less than a preset threshold, the first estimate is updated to the third estimate, and the looping steps are skipped; if the difference is less than the preset threshold, the third estimate is determined as the second estimate.

[0030] In some embodiments, the third estimate is an intermediate optimized value generated in each iteration of the gradient ascent method. When calculating the third estimate, it is necessary to use the first estimate of the current iteration as a basis, combine multiple valid sequences to weight and correct the first statistic of the current iteration, and then calculate the ratio between the updated first statistic and the fixed second statistic. The third estimate serves as a transitional estimate to determine whether the iteration has converged. Its core function is to gradually correct the deviation of the first estimate and push the result closer to the optimal solution.

[0031] It should be noted that the preset threshold is used to determine whether the third estimate has converged. The preset threshold can be set according to the actual situation. If more accurate calculation is required, the preset threshold can be set smaller. If the amount of calculation needs to be reduced, the preset threshold can be set larger.

[0032] In one possible implementation, the specific processing method of the above loop step is as follows: based on multiple valid sequences, calculate the fourth estimate corresponding to the first device, the (t+x)th device, and the preset number of device intervals in each valid sequence; where x is a positive integer and x is less than or equal to k; input all the fourth estimates, the first estimates, and the first statistic into the first update formula to obtain the updated first statistic; and determine the ratio between the updated first statistic and the second statistic as the third estimate.

[0033] In some embodiments, x takes values ​​from 1 to k, meaning that when k > 1, there will be multiple fourth estimates. The calculation method for the fourth estimate is the same as that for the first estimate before iterative optimization. The first update formula is:

[0034] in, This is the first statistic after the update. This is the first statistic before the update. For sequence The first estimate in, For sequence The fourth estimate in the data.

[0035] The following is a complete example to illustrate the solution of this application: The first step is to assume that there are n sequences of equipment power outage events in the historical data, and denote the m-th sequence of equipment power outage events as S. m S m (1) indicates the first equipment power outage event in the sequence, that is, the first equipment to malfunction, S m (t) represents the t-th device power outage event that occurs in the sequence, which is the t-th device that malfunctions.

[0036] The second step is to calculate the first statistic. Taking the m-th equipment power outage event sequence as an example, if the m-th equipment power outage event sequence S m The sequence includes a first device i, which is the h-th device in the sequence. The preset device interval is k, and S... m (h+k) represents the second device j. At this time, the m-th device power outage event sequence is the first target sequence. After traversing all device power outage event sequences, the total number of the first target sequences is the first statistic.

[0037] Third step, second statistic Taking the m-th equipment power outage event sequence as an example, if the m-th equipment power outage event sequence S mThe sequence includes a first device i, which is the t-th device in the sequence. The preset device interval is k, and the m-th device power outage event sequence contains S. m If (h+k), then the m-th equipment power outage event sequence is the second target sequence. After traversing all equipment power outage event sequences, the total number of the second target sequences is the second statistic.

[0038] The fourth step is to calculate the first estimated value. The first estimated value corresponding to the first device i, the second device j, and the preset device interval number k is... .

[0039] The fifth step is to filter out valid sequences from all the equipment power outage event sequences. Taking the m-th equipment power outage event sequence as an example, if the m-th equipment power outage event sequence S... m The sequence includes a first device i, which is the t-th device in the sequence. The preset device interval is k, and the m-th device power outage event sequence contains S. m If (t+k), then the m-th equipment power outage event sequence is a valid target sequence.

[0040] Step 6: Calculate the first statistic. Initialize to 0 ( =0).

[0041] Step 7: Calculate multiple fourth estimates based on all valid target sequences. The first statistic, the first estimate, and the fourth estimate are input into the first update formula. In the middle, the first statistic will be The value is updated to The third estimate is calculated using the updated first and second statistics. .

[0042] Step 8: Compare with the first estimate and the third estimate If the difference between the two is greater than a preset threshold, then let Proceed to step seven. If the difference between the two values ​​is not greater than a preset threshold, then the third estimated value will be... As a second estimate ( ).

[0043] In one possible implementation, after determining the impact intensity of the time-delay fault of the first device on the second device under a preset device interval number, the method further includes: constructing a similarity matrix under the preset device interval number based on the impact intensity of the time-delay fault of any two devices under the preset device interval number; solving for the eigenvalues ​​of the similarity matrix to obtain the eigenvectors of the similarity matrix; selecting the device corresponding to the row index of the largest number of elements of the preset device interval number from the eigenvectors of the similarity matrix, and determining the selected device as the key monitoring device under the preset device interval number, and notifying the dispatch control personnel to maintain the key monitoring device.

[0044] In some embodiments, when W represents a similarity matrix, the eigenvalues ​​of the similarity matrix W are solved by calculating the eigenvalue sequence of the similarity matrix W, which is equivalent to solving the determinant. The eigenvalues ​​are used to obtain the eigenvector corresponding to the second smallest eigenvalue. A=[a 1 ,a 2 ,...,a k ,...a N ] T Furthermore, the devices corresponding to the row indices of the k elements (preset device intervals) with the largest values ​​in feature vector A are identified as key monitoring devices. These key monitoring devices are then output to dispatch and control personnel as equipment objects that require important maintenance considerations in the future. The smaller the value of k, the more dangerous these devices are, and the higher the necessity for maintenance. For example, when k=1, devices 1, 2, and 3 are selected. This indicates that there is a high probability that the next device to fail in the power grid will be device 1, 2, or 3. Therefore, devices 1, 2, and 3 need to be maintained in a timely manner.

[0045] In one possible implementation, the specific processing method for "constructing a similarity matrix under a preset number of equipment intervals based on the time-delay fault influence intensity of any two devices under a preset number of equipment intervals" is as follows: construct a directed graph corresponding to the preset number of equipment intervals based on the time-delay fault influence intensity of any two devices under a preset number of equipment intervals; wherein, the length of the edge between the i-th node and the j-th node in the directed graph is the projected length of the time-delay fault influence intensity of any two devices under the preset number of equipment intervals between the i-th device and the j-th device in the power grid; the projected length is the reciprocal of the sum of the time-delay fault influence intensity and 0.01; the length of the edge between the i-th node and the j-th node in the directed graph is determined as the element in the i-th row and j-th column of the similarity matrix to obtain the similarity matrix.

[0046] In some embodiments, since the analysis of the impact intensity of time-delay faults needs to be performed for each preset number of equipment intervals, the analysis of the impact intensity of time-delay faults can be achieved by slicing the time-delay probability parameter tensor, fixing each preset number of equipment intervals, and obtaining multiple slice matrices.

[0047] It should be noted that when there are N devices in the power grid, the slice matrix... K The dimension is N×N, at this time, the slice matrix K In the diagram, the element with row coordinate i and ordinate j has a value of... K i,j = i,j,k This slice matrix represents the mutual influence relationships of all devices under a fixed preset device interval number k, for example, K i,j This represents the intensity of the time-delayed fault impact of the i-th device on the j-th device under a fixed preset number of device intervals, while K j,i This represents the intensity of the time-delay fault impact of the j-th device on the i-th device under a fixed preset number of device intervals.

[0048] In some embodiments, after obtaining the slice matrix, the time delay effect accumulation method of minimum projection path is used to calculate the shortest path from the first device to the second device under a fixed preset number of device intervals k. The shorter the path, the higher the intensity of the time delay fault effect of the first device on the second device.

[0049] The process of constructing the similarity matrix is ​​as follows: First, for each fixed preset number of device intervals... k ( k = 1, 2, ..., k (max), construct a directed graph G k , G k There are N nodes in total, where the length of the edge from the i-th node to the j-th node can be defined as... i,j,k The projected length, i,j,k The formula for calculating the projected length is 1 / ( K i,j +0.01). Secondly, for each fixed preset equipment interval number kConstruct a similarity matrix where the off-diagonal elements represent the minimum length between the i-th and j-th nodes in the directed graph. This can be calculated using a minimum path algorithm, such as a classic graph search algorithm (e.g., Dijkstra's algorithm). The formula for calculating the diagonal elements of the similarity matrix is:

[0050] Based on a sequence of multiple equipment outage events formed from historical equipment outage data of the power grid, statistical analysis is performed to obtain a first statistic corresponding to the first equipment, the second equipment, and a preset number of equipment intervals, as well as a second statistic corresponding to the first equipment and the preset number of equipment intervals. A first estimate is calculated by the ratio of these two types of statistics. This first estimate accurately characterizes the baseline time-delay impact of a first equipment failure on a second equipment failure after a preset number of intervals. Based on this, a gradient ascent method is used to iteratively optimize the first estimate, yielding a second estimate. This process further incorporates the correlation effects of other equipment failures during the process of the first equipment failure affecting the second equipment after a preset number of intervals, effectively improving the accuracy of the estimate and thus enhancing the reliability of the time-delay fault impact assessment results. Ultimately, this provides power grid dispatchers with a comprehensive and scientific basis for fault risk assessment, helping the power grid strengthen its fault resilience and ensure the safe and stable operation of the system.

[0051] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0052] The following are device embodiments of the present invention. For details not described in detail, please refer to the corresponding method embodiments described above.

[0053] Figure 3 A schematic diagram of the structure of the calculation device for the time-delay effect intensity of complex power grid faults provided in an embodiment of the present invention is shown. For ease of explanation, only the parts related to the embodiment of the present invention are shown, and are described in detail below: like Figure 3 As shown, the calculation device 3 for the intensity of the time-delay effect of complex power grid faults includes: The determination module 31 is used to determine the first statistic corresponding to the first device, the second device, and the preset device interval number in multiple device outage event sequences, and the second statistic corresponding to the first device and the preset device interval number; wherein, the preset device interval number is the number of devices in the power grid that fail after the first device fails, and the second device is the device in the power grid that fails after the first device fails, after a preset device interval number of devices. The calculation module 32 is used to determine the ratio of the first statistic and the second statistic as a first estimated value corresponding to the first device, the second device and the preset device interval number; The update module 33 is used to iteratively update the first estimate using the gradient ascent method and multiple equipment outage event sequences to obtain the second estimate corresponding to the first equipment, the second equipment, and the preset equipment interval number, and to use it as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset equipment interval number.

[0054] In one possible implementation, the determining module 31 is specifically used to: for each equipment power outage event sequence, if there is fault information of the first equipment in the equipment power outage event sequence, and the second equipment fails after the first equipment fails for a preset number of equipment intervals, then the equipment power outage event sequence is determined as the first target sequence; and the number of the first target sequence is determined as the first statistic.

[0055] In one possible implementation, the determining module 31 is further configured to: for each equipment power outage event sequence, if there is fault information of the first equipment in the equipment power outage event sequence, and after the first equipment fault preset equipment interval number, there is a equipment fault in the power grid, then the equipment power outage event sequence is determined as the second target sequence; and the number of the second target sequence is determined as the second statistic.

[0056] In one possible implementation, the update module 33 is specifically used to: for each equipment power outage event sequence, if the t-th faulty device in the equipment power outage event sequence is the first device and the (t+k)-th faulty device in the equipment power outage event sequence is the second device, then the equipment power outage event sequence is determined as a valid sequence; where t is a positive integer and k is a preset number of equipment intervals; based on multiple valid sequences and the second statistic, the first estimate is iteratively updated using the gradient ascent method to obtain the second estimate.

[0057] In one possible implementation, the update module 33 is further configured to: initialize the first statistic to 0 before iteration; loop step: update the first statistic based on multiple valid sequences and the first estimate, and update the first estimate based on the updated first and second statistics to obtain the third estimate; calculate the difference between the third estimate and the first estimate; if the difference is not less than a preset threshold, update the first estimate to the third estimate and jump to the loop step; if the difference is less than the preset threshold, determine the third estimate as the second estimate.

[0058] In one possible implementation, the update module 33 is further configured to: calculate, based on multiple valid sequences, a fourth estimate corresponding to the first device, the (t+x)th device, and the preset number of device intervals in each valid sequence; where x is a positive integer and x is less than or equal to k; input all the fourth estimates, the first estimates, and the first statistic into the first update formula to obtain the updated first statistic; and determine the ratio between the updated first statistic and the second statistic as the third estimate.

[0059] In one possible implementation, the update module 33 is further configured to: construct a similarity matrix under the preset equipment interval based on the time-delay fault impact intensity of any two devices under the preset equipment interval; solve for the eigenvalues ​​of the similarity matrix to obtain the eigenvector of the similarity matrix; select the device corresponding to the row index of the largest number of elements of the preset equipment interval from the eigenvector of the similarity matrix, and determine the selected device as the key monitoring device under the preset equipment interval, and notify the dispatch control personnel to maintain the key monitoring device.

[0060] In one possible implementation, the update module 33 is further configured to: construct a directed graph corresponding to the preset number of equipment intervals based on the time-delay fault impact intensity of any two devices under the preset number of equipment intervals; wherein, the length of the edge between the i-th node and the j-th node in the directed graph is the projected length of the time-delay fault impact intensity of any two devices between the i-th device and the j-th device in the power grid under the preset number of equipment intervals; the projected length is the reciprocal of the sum of the time-delay fault impact intensity and 0.01; and determine the length of the edge between the i-th node and the j-th node in the directed graph as the element in the i-th row and j-th column of the similarity matrix to obtain the similarity matrix.

[0061] Figure 4 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. For example... Figure 4 As shown, the electronic device 4 in this embodiment includes a processor 40 and a memory 41. The memory 41 stores a computer program 42. When the processor 40 executes the computer program 42, it implements the steps in the various method embodiments described above. Alternatively, when the processor 40 executes the computer program 42, it implements the functions of each module / unit in the various device embodiments described above.

[0062] For example, computer program 42 may be divided into one or more modules / units, which are stored in memory 41 and executed by processor 40 to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of computer program 42 in electronic device 4.

[0063] Electronic device 4 may include, but is not limited to, processor 40 and memory 41. Those skilled in the art will understand that... Figure 4 This is merely an example of electronic device 4 and does not constitute a limitation on electronic device 4. It may include more or fewer components than shown, or combine certain components, or different components. For example, electronic device 4 may also include input / output devices, network access devices, buses, etc.

[0064] For the sake of simplicity and clarity, only the above-described functional modules / units are used as examples. In practical applications, the functions described above can be assigned to different functional modules / units as needed. These modules / units can be implemented in hardware, software, or a combination of both.

[0065] In the above embodiments, the descriptions of each embodiment have their own emphasis. Parts not detailed or described in a particular embodiment can be referred to in the relevant descriptions of other embodiments. Unless otherwise specified or in conflict with logic, the terminology and / or descriptions between different embodiments are consistent and can be referenced interchangeably. Technical features in different embodiments can be combined to form new embodiments based on their inherent logical relationships.

[0066] The above-described embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A method for calculating the intensity of the time-delay effect of faults in complex power grids, characterized in that, include: In a sequence of multiple equipment outage events, determine the first statistic corresponding to the first equipment, the second equipment, and the preset equipment interval number, and the second statistic corresponding to the first equipment and the preset equipment interval number; wherein, the preset equipment interval number is the number of equipment in the power grid that fails after the first equipment fails, and the second equipment is the equipment in the power grid that fails at the preset equipment interval number after the first equipment fails; The ratio of the first statistic to the second statistic is determined as the first estimated value corresponding to the number of intervals between the first device, the second device, and the preset device. Using the gradient ascent method and multiple equipment outage event sequences, the first estimated value is iteratively updated to obtain the second estimated value corresponding to the first equipment, the second equipment, and the preset equipment interval number. This second estimated value is used as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset equipment interval number.

2. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 1, characterized in that, The preset device interval number is a positive integer; the step of iteratively updating the first estimate using the gradient ascent method and multiple device power outage event sequences to obtain the second estimate corresponding to the first device, the second device, and the preset device interval number includes: For each equipment power outage event sequence, if the t-th faulty device in the equipment power outage event sequence is the first device, and the (t+k)-th faulty device in the equipment power outage event sequence is the second device, then the equipment power outage event sequence is determined to be a valid sequence. Where t is a positive integer; k is the preset number of device intervals; Based on multiple valid sequences and the second statistic, the first estimate is iteratively updated using the gradient ascent method to obtain the second estimate.

3. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 2, characterized in that, The step of iteratively updating the first estimate based on multiple valid sequences and the second statistic using the gradient ascent method to obtain the second estimate includes: Before the iteration, the first statistic is initialized to 0; The iterative steps are as follows: based on multiple valid sequences and the first estimated value, the first statistic is updated, and based on the updated first statistic and the second statistic, the first estimated value is updated to obtain the third estimated value. Calculate the difference between the third estimate and the first estimate; If the difference is not less than a preset threshold, the first estimated value is updated to the third estimated value, and the process jumps to the loop step. If the difference is less than a preset threshold, then the third estimated value is determined as the second estimated value.

4. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 3, characterized in that, The step of updating the first statistic based on multiple valid sequences and the first estimated value, and updating the first estimated value based on the updated first statistic and the second statistic to obtain the third estimated value, includes: Based on multiple valid sequences, calculate the fourth estimate corresponding to the first device, the (t+x)th device, and the preset number of device intervals in each valid sequence; where x is a positive integer and x is less than or equal to k; Input all the fourth estimates, the first estimates, and the first statistic into the first update formula to obtain the updated first statistic; The ratio between the updated first statistic and the second statistic is determined as the third estimate.

5. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 1, characterized in that, The first statistic corresponding to the first device, the second device, and the preset device interval number is determined, including: For each equipment power outage event sequence, if the equipment power outage event sequence contains fault information of the first equipment, and the second equipment fails after a preset equipment interval number of the first equipment failure, then the equipment power outage event sequence is determined as the first target sequence; The number of the first target sequence is determined as the first statistic.

6. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 1, characterized in that, The second statistic, which determines the first device and the preset device interval number in a sequence of multiple device power outage events, includes: For each equipment outage event sequence, if the equipment outage event sequence contains fault information of the first equipment, and after the first equipment fault has a preset equipment interval number, the power grid has a equipment fault, then the equipment outage event sequence is determined as the second target sequence. The number of the second target sequence is determined as the second statistic.

7. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 1, characterized in that, After determining the second estimated value corresponding to the first device, the second device, and the preset device interval number as the intensity of the time-delay fault impact of the first device on the second device below the preset device interval number, the method further includes: Based on the time-delay fault impact intensity of any two devices under the preset device interval number, construct a similarity matrix under the preset device interval number; Solve for the eigenvalues ​​of the similarity matrix to obtain the eigenvectors of the similarity matrix; In the eigenvector of the similarity matrix, select the device corresponding to the row index of the element with the largest preset device interval number, and determine the selected device as the key monitoring device below the preset device interval number, and notify the dispatch control personnel to maintain the key monitoring device.

8. The method for calculating the intensity of the time-delay effect of complex power grid faults according to claim 7, characterized in that, The step of constructing a similarity matrix based on the time-delay fault impact intensity of any two devices within the preset device interval number includes: Based on the impact intensity of time-delay faults between any two devices under the preset device interval number, construct a directed graph corresponding to the preset device interval number; Wherein, the length of the edge between the i-th node and the j-th node in the directed graph is the projected length of the time-delay fault influence intensity of any two devices between the i-th device and the j-th device in the power grid below the preset device interval number; the projected length is the reciprocal of the sum of the time-delay fault influence intensity and 0.01; The length of the edge between the i-th node and the j-th node in the directed graph is determined as the element in the i-th row and j-th column of the similarity matrix, thus obtaining the similarity matrix.

9. A calculation device for the intensity of the time-delay effect of complex power grid faults, characterized in that, include: The determination module is used to determine the first statistic corresponding to the first device, the second device, and the preset device interval number in a sequence of multiple device power outage events, and the second statistic corresponding to the first device and the preset device interval number; wherein, the preset device interval number is the number of devices in the power grid that fail after the first device fails, and the second device is the device in the power grid that fails after the first device fails, after a preset device interval number of devices; The calculation module is used to determine the ratio of the first statistic and the second statistic as a first estimated value corresponding to the number of intervals between the first device, the second device, and the preset device; The update module is used to iteratively update the first estimate using the gradient ascent method and multiple equipment outage event sequences to obtain the second estimate corresponding to the first equipment, the second equipment, and a preset number of equipment intervals, and to use it as the intensity of the time-delay fault impact of the first equipment on the second equipment under the preset number of equipment intervals.

10. An electronic device, characterized in that, It includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the method as described in any one of claims 1 to 8.

Images