Power distribution system scheduling method and device applied to extreme weather, equipment and medium

Abstract

The invention relates to a power distribution system scheduling method and device applied to extreme weather, equipment and a medium. The method comprises the steps of obtaining weather data and distribution line parameters; wherein the distribution line parameters comprise resistance strength parameters of a plurality of distribution lines; determining the fault probability of the plurality of distribution lines according to the weather data and the resistance strength parameters; according to the fault probability of each distribution line, simulating the fault scene for multiple times to obtain a plurality of line fault scene combinations; wherein each line fault scene combination comprises a predicted value of a fault of each distribution line; inputting the plurality of line fault scene combinations into the power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is obtained by training a fault prevention scheme and a fault processing scheme corresponding to various historical line fault scene combinations. By adopting the method, the power supply safety of the power distribution system in extreme weather can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent power distribution, in particular to a power distribution system scheduling method and device applied to extreme weather, equipment and medium. BACKGROUND

[0002] Under extreme weather, the lines and towers of the power distribution network may be broken or collapsed, which seriously affects the power supply reliability of the power distribution network. At present, the pre-disaster fault uncertainty is rarely considered in the research of power distribution network scheduling, and the fault uncertainty is not handled properly, which leads to the decline of the safe operation ability of the power distribution network and the reduction of the reliability under extreme ice and snow weather.

[0003] Although some research has proposed a power distribution network reconstruction method based on fault probability, most of them use a single-layer optimization structure, which is difficult to consider the decision-making needs of both the pre-disaster prevention and disaster response stages, resulting in insufficient overall response capacity of the system. For the generation and processing of line fault scenarios, related research usually uses deterministic methods or simple probability models, which fails to fully consider the relationship between meteorological data and fault probability, affecting the reliability of power distribution system scheduling and making the power supply safety of the power distribution system under extreme weather insufficient. SUMMARY

[0004] Therefore, it is necessary to provide a power distribution system scheduling method, device, equipment and medium applied to extreme weather, which can improve the power supply safety of the power distribution system under extreme weather.

[0005] In a first aspect, the present application provides a power distribution system scheduling method applied to extreme weather, comprising:

[0006] obtaining weather data and power distribution line parameters; wherein the power distribution line parameters include the resistance strength parameters of multiple power distribution lines;

[0007] determining the fault probability of each of the multiple power distribution lines according to the weather data and the resistance strength parameters;

[0008] According to the fault probability of each of the power distribution lines, multiple simulations of fault scenarios are performed to obtain multiple line fault scenario combinations; wherein each of the line fault scenario combinations includes the predicted value of the fault of each of the power distribution lines;

[0009] inputting the multiple line fault scenario combinations into a power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is trained according to the fault prevention schemes and fault handling schemes corresponding to multiple historical line fault scenario combinations.

[0010] In one of the embodiments, the power distribution scheduling optimization model comprises a fault prevention sub-model and a fault handling sub-model; the inputting of the plurality of line fault scenario combinations into the power distribution scheduling optimization model to obtain a power distribution scheduling scheme comprises:

[0011] The plurality of line fault scenario combinations are inputted into the fault prevention sub-model to obtain power distribution network topology and energy storage pre-layout data;

[0012] The power distribution network topology and the energy storage pre-layout data are inputted into the fault handling sub-model to obtain a power distribution scheduling scheme under each line fault scenario.

[0013] In one of the embodiments, after the plurality of line fault scenario combinations are obtained by the plurality of times of simulation of the fault scenarios according to the fault probabilities of the power distribution lines, the method further comprises:

[0014] The plurality of line fault scenario combinations are clustered to obtain a plurality of clustering clusters;

[0015] For each of the clustering clusters, a line fault scenario combination with a smallest distance value from a clustering center is selected as a target fault scenario combination;

[0016] The inputting of the plurality of line fault scenario combinations into the power distribution scheduling optimization model to obtain a power distribution scheduling scheme comprises:

[0017] The plurality of target fault scenario combinations are inputted into the power distribution scheduling optimization model to obtain a power distribution scheduling scheme.

[0018] In one of the embodiments, the clustering of the plurality of line fault scenario combinations to obtain a plurality of clustering clusters comprises:

[0019] A clustering parameter is obtained, and a preset number of line fault scenario combinations are selected as initial clustering centers according to the clustering parameter to form a plurality of initial clustering clusters;

[0020] For each of the line fault scenario combinations, a distance value of the line fault scenario combination from each of the initial clustering centers is calculated, and the line fault scenario combination is assigned to an initial clustering cluster corresponding to the smallest distance value;

[0021] In a case where all of the line fault scenario combinations have been assigned to each of the initial clustering clusters, each of the initial clustering centers is updated, and the step of calculating the distance value of the line fault scenario combination from each of the initial clustering centers is repeated until a preset stop condition is met, and a plurality of clustering clusters are obtained.

[0022] In one of the embodiments, the plurality of line fault scenario combinations are obtained by simulating a fault scenario multiple times according to the fault probability of each of the power distribution lines, including:

[0023] In each of the simulation of the scenario, a random number in a numerical range of the fault probability is generated for each of the power distribution lines;

[0024] In a case where the fault probability is greater than or equal to the random number, it is determined that the simulation result of the power distribution line in the simulation of the scenario is a fault;

[0025] In a case where the fault probability is less than the random number, it is determined that the simulation result of the power distribution line in the simulation of the scenario is no fault;

[0026] The simulation results of each of the power distribution lines in the simulation of the scenario are combined to obtain the line fault scenario combination.

[0027] In one of the embodiments, the fault probability of each of the plurality of power distribution lines is determined according to the weather data and the resistance intensity parameter, including:

[0028] An ice thickness prediction value is calculated according to the weather data;

[0029] The ice thickness prediction value and the resistance intensity parameter of each of the power distribution lines are input into a line fault prediction model to obtain the fault probability of each of the plurality of power distribution lines.

[0030] In a second aspect, the application further provides a power distribution system scheduling device for extreme weather, including:

[0031] A data acquisition module is configured to acquire weather data and power distribution line parameters; wherein the power distribution line parameters include resistance intensity parameters of a plurality of power distribution lines;

[0032] A fault analysis module is configured to determine fault probabilities of a plurality of power distribution lines according to the weather data and the resistance intensity parameters;

[0033] A fault prediction module is configured to simulate a fault scenario multiple times according to the fault probability of each of the power distribution lines to obtain a plurality of line fault scenario combinations; wherein each of the line fault scenario combinations includes a prediction value of a fault of each of the power distribution lines;

[0034] A scheduling module is configured to input the plurality of line fault scenario combinations into a power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is trained according to a plurality of historical line fault scenario combinations and corresponding fault prevention schemes and fault handling schemes.

[0035] In a third aspect, the present application also provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the first aspect when executing the computer program.

[0036] In a fourth aspect, the present application also provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the first aspect when executed by a processor.

[0037] In a fifth aspect, the present application also provides a computer program product comprising a computer program, and the computer program implements the steps of the first aspect when executed by a processor.

[0038] The above power distribution system scheduling method and device for extreme weather, computer device, computer readable storage medium and computer program product can obtain weather data and power distribution line parameters, wherein the power distribution line parameters comprise resistance strength parameters of multiple power distribution lines, the physical characteristics of the power distribution network and the current meteorological conditions faced can be comprehensively mastered, the failure probability of the multiple power distribution lines is determined according to the weather data and the resistance strength parameters, the potential influence of the extreme ice and snow weather on each power distribution line can be quantitatively evaluated, multiple simulations are performed on the failure scenarios according to the failure probability of each power distribution line, and multiple line failure scenario combinations are obtained, wherein each line failure scenario combination comprises a predicted value of failure of each power distribution line, a statistically representative failure scenario set can be generated through multiple simulations, the multiple line failure scenario combinations are input into the power distribution scheduling optimization model trained according to the failure prevention schemes and the failure handling schemes corresponding to multiple historical line failure scenario combinations, a power distribution scheduling scheme is obtained, the probability model and the multiple scenario simulations are used to comprehensively consider various possible failure combinations and their occurrence probabilities, and the power distribution scheduling optimization model considering the failure prevention schemes and the failure handling schemes is used to determine the scheduling scheme, so that the ability of the system to cope with the extreme ice and snow weather is improved, and the power supply safety of the power distribution system under the extreme weather is improved. BRIEF DESCRIPTION OF DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related art, the drawings needed to be used in the description of the embodiments of the present application or the related art will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0040] Figure 1 An application environment diagram of the power distribution system scheduling method for extreme weather in an embodiment;

[0041] Figure 2A flowchart of a power distribution system dispatching method applied to extreme weather in an embodiment;

[0042] Figure 3 A schematic diagram of a bi-level optimization model structure of a power distribution dispatching optimization model in an embodiment;

[0043] Figure 4 A flowchart of a power distribution system dispatching method applied to extreme weather in another embodiment;

[0044] Figure 5 A schematic diagram of a power distribution network simulation line in an embodiment;

[0045] Figure 6 A schematic diagram of an upper layer optimization result of a bi-level optimization model structure in an embodiment;

[0046] Figure 7 A schematic diagram of a lower layer optimization result of a bi-level optimization model structure in an embodiment;

[0047] Figure 8 A schematic diagram of an optimization result of a bi-level optimization model structure in an embodiment;

[0048] Figure 9 A block diagram of a power distribution system dispatching device applied to extreme weather in an embodiment;

[0049] Figure 10 An internal structure diagram of a computer device in an embodiment. DETAILED DESCRIPTION

[0050] In order to make the purpose, technical scheme and advantages of the present application clearer, the present application is further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0051] It should be noted that the terms "first", "second", etc. used in the present application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "include" and "have" and any variations thereof used in the present application are intended to cover non-exclusive inclusion. The term "a plurality of" used in the present application means two or more. The term "and / or" used in the present application means one of the options or any combination of multiple options.

[0052] The power distribution system dispatching method applied to extreme weather provided by the embodiments of the present application can be applied to, for example, Figure 1The application environment shown. Among them, the terminal 102 communicates with the server 104 through the network, the terminal 102 can be used to obtain weather data and power distribution line parameters and other data, and send the obtained data to the server 104 for analysis and processing. The data storage system can store the data required by the server 104 for processing. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. The data storage system can be used to store historical and current weather data and power distribution line parameters and other data. Among them, the terminal 102 can be, but not limited to, various personal computers, notebook computers, smart phones, tablet computers, unmanned aerial vehicles, low-altitude aircraft, Internet of Things devices and portable wearable devices, Internet of Things devices can be smart speakers, smart televisions, smart air conditioners, smart car devices, projection devices, etc. Portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc. The server 104 can be a standalone physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0053] In an exemplary embodiment, as Figure 2 shown, a power distribution system scheduling method for extreme weather is provided, and the method is applied to Figure 1 the server 104 in the above-mentioned application environment as an example for illustration, including the following steps S202 to S208. Among them:

[0054] Step S202, obtaining weather data and power distribution line parameters.

[0055] Among them, the power distribution line parameters include the resistance strength parameters of a plurality of power distribution lines. Exemplarily, the weather data can include but is not limited to meteorological parameters such as liquid water content, precipitation rate, wind speed, etc.; the power distribution line parameters can include the resistance strength parameters of a plurality of power distribution lines, such as the designed ice resistance strength of the line, the length of the line, the direction of the line, etc. The resistance strength parameter can be used to represent the bearing capacity of the power distribution line to the extreme weather. Exemplarily, the resistance strength parameter can be the ice resistance strength of the power distribution line, for example, representing the maximum ice thickness that the power distribution line can withstand, which can be in millimeters.

[0056] Exemplarily, the server 104 can obtain real-time or forecast weather data through the data interface with the meteorological department, and obtain the power distribution line parameters from the database of the power distribution network management system. For example, the server 104 can obtain the weather forecast data of a certain area for the next 48 hours, including the liquid water content of 0.5g / m 3information such as an average precipitation rate of 2 mm / h, an average wind speed of 5 m / s, and the like; and meanwhile, the design ice resistance of each line in the power distribution network in the region is obtained, such as that the design ice resistance of line L1 is 10 mm, and the design ice resistance of line L2 is 12 mm, and the like.

[0057] In step S204, the failure probabilities of the plurality of power distribution lines are determined according to the weather data and the resistance strength parameters.

[0058] For example, the server 104 can calculate an ice thickness prediction value according to the weather data, and input the ice thickness prediction value and the resistance strength parameters of each power distribution line into a line failure prediction model to obtain the failure probabilities of the plurality of power distribution lines.

[0059] The line failure prediction model refers to a mathematical model established according to the relationship between the line ice thickness and the failure probability, and can be used to calculate the failure probability of the line under different ice thicknesses. The ice thickness prediction value refers to the ice thickness of the power distribution line predicted according to the weather data, and the unit can be millimeters. The failure probability refers to the probability value of the failure of the power distribution line under the extreme ice and snow weather, and the range is a value between 0 and 1.

[0060] For example, the server 104 calculates the ice thickness prediction value by using a numerical calculation model of ice prediction, and can construct a vulnerability model of the power distribution line as the line failure prediction model, which can reflect the relationship between the line ice thickness and the failure probability. The ice thickness prediction value and the resistance strength parameters of each power distribution line are input into the line failure prediction model to obtain the failure probabilities of the plurality of power distribution lines. For example, when the ice thickness is lower than the design ice resistance of the conductor, the line failure probability is very small and close to 0; when the ice thickness is between the design ice resistance of the conductor and 5 times the design ice resistance, the failure probability increases linearly with the ice thickness; and when the ice thickness reaches 5 times the design ice resistance, the line failure probability is 1, that is, the failure is inevitable.

[0061] The numerical calculation model of ice prediction and the vulnerability model of the power distribution line are constructed, and the numerical calculation model of ice prediction is as follows:

[0062]

[0063] wherein, R it is the ice thickness of the conductor at time t, W j is the liquid water content in the air, N represents the ice duration at time t, j represents the jth hour of the ice duration, P j is the precipitation rate, v j is the wind speed; p i is the ice density, and 0.9 g / cm3 is taken; p oWater density is taken as 1 g / cm3. For example, for a certain distribution line, assuming ice accumulation lasts for 24 hours, the average liquid water content is 0.5 g / m 3 , the average precipitation rate is 2 mm / h, the average wind speed is 5 m / s, the ice density is 0.9 g / cm 3 , and the water density is 1 g / cm 3 , the server 104 can calculate that the ice thickness prediction value of the line is about 8.93 mm.

[0064] The distribution line vulnerability model is as follows:

[0065]

[0066] wherein, let be the failure probability of the line ij; p nor be the failure probability when the ice thickness is lower than the designed ice resistance strength of the conductor, at this time the line failure probability is very small, close to 0; a be the designed ice resistance strength of the conductor, b be 5 times of the designed ice resistance strength, unit mm, when the ice thickness reaches 5 times of the designed ice resistance strength, the line failure probability is 1.

[0067] Through the above steps, the server 104 can calculate a failure probability value for each distribution line in the system. For example, for a distribution network containing 5 main lines, the server 104 can calculate the following failure probabilities: the failure probability of line L5 is 0.56, the failure probability of line L6 is 0.55, the failure probability of line L19 is 0.53, the failure probability of line L26 is 0.56, and the failure probability of line L29 is 0.57. These failure probabilities reflect the possibility of failure of each line under extreme ice and snow weather, and the larger the value, the higher the risk of line failure.

[0068] The distribution line vulnerability model constructed by the server 104 considers the nonlinear relationship between ice thickness and failure probability, and can more accurately reflect the vulnerability characteristics of the line under different ice conditions. Compared with the traditional binary judgment method (i.e. the line will certainly fail when the ice thickness exceeds the threshold, otherwise it will not fail), this probabilistic processing method is more in line with the actual situation, and can more finely describe the failure risk. For example, for a line with a designed ice resistance strength of 15 mm, the traditional method may simply judge that the line will fail when the ice thickness exceeds 15 mm, while the vulnerability model adopted by the server 104 can give a failure probability that gradually increases with the increase of ice thickness, such as a failure probability of 0.14 when the ice thickness is 20 mm, a failure probability of 0.43 when the ice thickness is 30 mm, and a failure probability of 0.71 when the ice thickness is 40 mm.

[0069] Further, the server 104 can also consider more factors that affect line failure to build a more complex failure probability model. For example, the server 104 can take into account factors such as conductor material properties, service life, historical failure records, etc., to modify the base failure probability model. For a line with a long service life, the server 104 can increase its base failure probability; for a line that has failed in similar weather conditions in the past, the server 104 can also increase its failure probability assessment value. In addition, the server 104 can also consider the geographical environment in which the line is located, such as a line in a mountainous area may be more susceptible to the combined effects of icing and strong winds than a line in a plain, and the risk of failure is higher. For example, for a line located in a mountainous area with a service life of 15 years, even if its icing thickness prediction value does not exceed the designed ice resistance strength, the server 104 can adjust its failure probability from close to 0 to 0.05 or higher to reflect its overall risk level.

[0070] In step S206, a plurality of line failure scenario combinations are obtained by simulating the failure scenarios according to the failure probabilities of the respective distribution lines.

[0071] Each line failure scenario combination includes the predicted values of the respective distribution lines failing.

[0072] Exemplarily, in each scenario simulation, the server 104 can generate a random number within the numerical range of the failure probability for each distribution line; in the case where the failure probability is greater than or equal to the random number, it is determined that the simulated result of the distribution line in the scenario simulation is failure; in the case where the failure probability is less than the random number, it is determined that the simulated result of the distribution line in the scenario simulation is non-failure; the simulated results of the respective distribution lines in the scenario simulation are combined to obtain a line failure scenario combination. Exemplarily, the server 104 can use the Monte Carlo simulation method to generate a statistically representative set of failure scenarios by means of random sampling to reflect the randomness and complexity of distribution line failure under extreme ice and snow weather. The server 104 can build a failure scenario generation model for the distribution lines. In this model, the distribution lines have two states, failure or non-failure, and let L fail be the line state variable, then L fail is 0 when it fails, and L fail is 1 when it does not fail. According to the line failure probability, a large number of failure scenarios can be generated by Monte Carlo simulation. When the random number cr is not higher than the failure probability of line i, it is considered that line i fails, otherwise it does not fail:

[0073]

[0074] where cr is a uniformly distributed random number; Ω line is the set of all distribution lines.

[0075] In each simulation, the server 104 can generate a random number within the numerical range of the failure probability for each distribution line. These random numbers are uniformly distributed between 0 and 1. Then, the server 104 can compare the random number with the failure probability of the line: in the case where the failure probability is greater than or equal to the random number, it is determined that the simulation result of the distribution line in the scenario simulation is a failure; in the case where the failure probability is less than the random number, it is determined that the simulation result of the distribution line in the scenario simulation is not a failure. Finally, the server 104 combines the simulation results of each distribution line in the scenario simulation to obtain a complete line failure scenario combination.

[0076] For example, for a distribution network containing 5 main lines (L5, L6, L19, L26, L29), suppose the failure probabilities of these lines are 0.56, 0.55, 0.53, 0.56, 0.57 respectively. In the first simulation, the server 104 can generate 5 random numbers for these 5 lines: [0.62, 0.48, 0.71, 0.34, 0.82]. Comparing the random numbers with the failure probabilities: line L5: 0.62>0.56, the simulation result is not a failure; line L6: 0.48<0.55, the simulation result is a failure; line L19: 0.71>0.53, the simulation result is not a failure; line L26: 0.34<0.56, the simulation result is a failure; line L29: 0.82>0.57, the simulation result is not a failure. Therefore, the failure scenario combination obtained in the first simulation is that lines L6 and L26 fail, and the remaining lines are normal. In the second simulation, the server 104 can generate different random numbers, such as [0.23, 0.68, 0.22, 0.93, 0.33]. After comparison, the simulation result is that lines L5, L19 and L29 fail, and lines L6 and L26 are normal. This constitutes a second failure scenario combination.

[0077] By repeating the above process multiple times, the server 104 can generate hundreds or thousands of different failure scenario combinations. Each failure scenario combination represents a possible line failure combination, which contains the predicted value of whether each distribution line fails (0 represents a failure, and 1 represents no failure). These failure scenario combinations together constitute a scenario set.

[0078] Step S208, inputting the multiple line failure scenario combinations into the distribution dispatching optimization model to obtain a distribution dispatching scheme.

[0079] The power distribution dispatch optimization model is trained according to the corresponding fault prevention scheme and fault handling scheme of a plurality of historical line fault scenarios. The power distribution dispatch optimization model can generate an optimal power distribution dispatch scheme according to the input fault scenario combination, so as to maximize the power supply reliability and reduce the economic loss. The server 104 can collect a large amount of historical fault data, which can include fault scenarios, preventive measures (such as power distribution network reconstruction, mobile energy storage pre-deployment, etc.), coping strategies (such as mobile energy storage dispatching, diesel generator starting, etc.), and final effects (such as load recovery rate, economic loss, etc.). Then, the server 104 can train the optimization model using these data, so that it can automatically generate an effective dispatching scheme according to a new fault scenario.

[0080] Referring to Figure 3 The power distribution dispatch optimization model can be a double-layer optimization model structure including a fault prevention sub-model and a fault handling sub-model. This double-layer structure can reflect two stages of decision-making: pre-disaster prevention (upper layer) and disaster response (lower layer), and can simultaneously consider the decision-making needs of the two stages of pre-disaster prevention and disaster response. Illustratively, the server 104 can input a plurality of line fault scenario combinations into the fault prevention sub-model to obtain power distribution network topology and energy storage pre-deployment data; and input the power distribution network topology and energy storage pre-deployment data into the fault handling sub-model to obtain a power distribution dispatch scheme under each line fault scenario.

[0081] The optimization objective of the fault prevention sub-model (upper layer model) is to minimize the mobile energy storage pre-deployment cost and the worst-case loss of load cost. This model can consider all possible fault scenarios, develop a robust initial scheme, including how the power distribution network should be reconstructed (for example, which switches should be opened and which should be closed), and how the mobile energy storage should be pre-deployed (for example, how many mobile energy storages, and where to pre-deploy), etc. For example, for the 10 fault scenarios generated in the above steps, the server 104 can obtain the optimization result of the fault prevention sub-model as follows: disconnect the connection between lines L8-L12 and lines L20-L25, and pre-deploy 1 mobile energy storage at node N3 and node N7.

[0082] Subsequently, the server 104 can input the power distribution network topology and energy storage pre-layout data into a fault handling sub-model to obtain a power distribution scheduling scheme under each line fault scenario. The optimization objective of the fault handling sub-model (lower-level model) is to minimize the loss of load cost under a specific fault scenario. This model can develop a detailed scheduling scheme for each specific fault scenario based on the determined network topology and energy storage pre-layout, including dynamic scheduling of mobile energy storage (e.g., when and where to move, when to charge and discharge), diesel generator output control, load reduction strategy, etc. For example, for fault scenario 7 (lines L5, L19, and L29 simultaneously fault), the server 104 can obtain a scheduling scheme through the fault handling sub-model: move energy storage 1 from node N3 to node N5 at t = 2, and start discharging at t = 4; move energy storage 2 from node N7 to node N15 at t = 1, and start discharging at t = 3; start the diesel generator at node N10 at t = 1, with an output power of 80% of the rated value; temporarily reduce the non-essential load at nodes N18 and N23, and restore power supply after the energy storage is in place.

[0083] Further, the power distribution network multi-element collaborative double-layer optimization model constructed by the server 104 contains multiple constraint conditions, ensuring that the generated scheduling scheme is feasible and reasonable. These constraints include radial topology constraints (to ensure that the power distribution network maintains a radial structure), power distribution network operation constraints (such as power flow balance, voltage limitation, etc.), mobile energy storage dynamic scheduling constraints considering traffic network integration (considering the impact of extreme weather on the driving speed of mobile energy storage), diesel generator constraints (such as upper and lower output limits, climbing constraints, etc.), slack bus constraints, on-load tap-changing transformer constraints, group switching capacitor constraints, static var compensator constraints, etc. For example, for the scheduling of mobile energy storage, the server 104 can consider the impact of extreme snow and ice weather on the traffic network, calculate the actual driving speed (usually lower than the ideal speed) and arrival time of the mobile energy storage, and ensure the feasibility of the scheduling scheme.

[0084] For example, the radial topology constraint is as follows: the power distribution network needs to satisfy the radial topology constraint, so the concepts of "virtual node" and "virtual power" are introduced:

[0085]

[0086] where α ij is a line connection state 0-1 variable, with a value of 1 indicating that node i is connected to j, and a value of 0 indicating that it is disconnected; N B is the total number of nodes; S i is an introduced virtual node state 0-1 variable; Ω B represents a branch set, Ω N represents a virtual node set; F i represents virtual node power; Fij Let represent the virtual branch power; j∈δ(i) represents the set of all branches with i as the first node, and k∈γ(i) represents the set of all branches with i as the last node. The virtual power is relaxed using the Big M method, i.e.:

[0087] -M1S i ≤F i ≤M1S i

[0088] -M1α ij ≤F ij ≤M1α ij

[0089] M1 is a very large positive number.

[0090] Furthermore, the constraints also include distribution network operation constraints. Distribution network operation needs to satisfy power flow balance, and the active power flow balance constraint is shown in the following formula:

[0091]

[0092] Among them, P ij and P jk P represents the transmitted active power of branches ij and jk, respectively; j Let v(j) represent the active power of node j, and v(j) represent all branches starting from node j; r ij The resistance of branch ij is represented by l. ij P is the square of the current in branch ij; load,j P represents the active power of the load at node j. 0,j P represents the active power at the slack node. PV,j P represents photovoltaic output. DEG,j P represents the output power of the diesel generator at node j. dis,j P represents the energy storage charging power of node j. ch,j This represents the energy storage and discharge power of node j; This indicates that the active power reduction of the load at node j is... Low temperatures and rain or snow often cause a decrease in photovoltaic power output, and the calculation formula is as follows:

[0093]

[0094] in, For photovoltaic output under ideal conditions, l di Let k be the daily energy loss rate, h be the fitting coefficient, and k be the daily energy loss rate. snow θ is the monthly snowfall, RH is the tilt angle of the photovoltaic panel, and T is the air humidity. mis the monthly average temperature, HG is the irradiance, R is the width of the photovoltaic array in the inclined plane, H is the height of the photovoltaic panel, q is the avalanche slope angle, L is the monthly energy loss rate, month is the number of days in the month, is the energy estimate value of the di day, is the introduced snowfall influence factor, l0 is an unknown to be solved.

[0095] Correspondingly, the reactive power flow balance constraint is as follows:

[0096]

[0097] Where, Q ij and Q jk are the transmission reactive power of branch ij and branch jk respectively; Q j is the reactive power of node j; x ij represents the reactance of branch ij; Q load,j represents the load reactive power of node j, Q CB,j represents the group switching capacitor group power at node j, Q SVC,j represents the static reactive power compensator power at node j, Q 0,j represents the balance node active power; represents the load reduction reactive power of node j, Let Vi and Vj be the voltage square of node i and node j, then:

[0098]

[0099] Where, M1 is a very large positive number, is the square of the minimum voltage, is the square of the maximum voltage.

[0100] Further, the server 104 can also consider the traffic network integrated mobile energy storage dynamic scheduling model when establishing the model, and the preconfigured mobile energy storage cannot exceed its upper limit, that is:

[0101]

[0102] Where, Ω N is the load node set, is the connection state variable of the mobile energy storage and node i, is the upper limit of the number of mobile energy storages. Under disaster weather, the driving speed of the mobile energy storage is affected, and the equivalent passing time, equivalent distance and actual driving speed calculation formula are as follows:

[0103]

[0104] Where, D j,k(t) represents the equivalent travel distance from node j to k, D j,k,0 represents the actual distance; represents the actual vehicle speed of mobile energy storage considering the impact of ice disaster, v i,0 represents the vehicle speed under ideal conditions, c is the introduced disaster impact factor; represents the equivalent travel time of mobile energy storage. Similar to fixed energy storage, mobile energy storage also needs to meet the charge and discharge power constraints shown in the following formula:

[0105] E s,t = E s,t-1 + η ch P ch,t - P dis,t / η dis

[0106]

[0107] E s,min ≤ E s,t ≤ E s,max

[0108] When a disaster occurs, the mobile energy storage should be located at the pre-layout position, that is,

[0109]

[0110] During the process of mobile energy storage moving from node j to node k, it is not connected to node k, that is,

[0111]

[0112] The same mobile energy storage cannot be connected to two or more nodes at time t, that is,

[0113]

[0114] The mobile energy storage cannot be in the state of charging and discharging at the same time, that is,

[0115]

[0116] wherein E s,t represents the energy of the energy storage at time t, E s,t-1 represents the energy at the previous time; P ch,t represents the charging power of the energy storage at time t, η ch represents the charging efficiency; P dis,t represents the discharging power of the energy storage at time t, η dis represents the discharging efficiency; represents the charging state of the mobile energy storage, represents the discharging state; E s,min represents the lower limit of the capacity of the mobile energy storage, E s,maxrepresents the upper limit of capacity; t0 represents the time when the disaster occurs; represents the connection state of the mobile energy storage and node j at the time when the disaster occurs; represents the installation configuration time of the mobile energy storage; represents the connection state of the mobile energy storage and node j at time t, represents the connection state at the next time.

[0117] Further, in addition to meeting the upper and lower limit constraints of the output, the diesel generator also needs to meet the climbing constraint, that is, the power difference between the current time and the previous time cannot exceed the limit, and the constraint is as follows:

[0118] P DEG,min ≤P DEG,j ≤P DEG,max

[0119]

[0120] wherein, P DEG,j represents the power of the diesel generator, P DEG,max represents the maximum power, P DEG,min represents the minimum power, R Ui is the climbing rate, R Di is the sliding rate, and represent the power at time t and the previous time, respectively.

[0121] Further, the server 104 can select the transformer substation as the balancing node, and the output thereof needs to meet the upper and lower limit constraints, as follows:

[0122]

[0123] wherein, P0 represents the active power of the balancing node, Q0 represents the reactive power of the balancing node, P 0,max and Q 0,max represent the maximum active and reactive power thereof, P 0,min and Q 0,min represent the minimum power thereof.

[0124] The active power grid can adjust its voltage level through the on-load voltage regulating transformer, as shown in the following formula:

[0125]

[0126] Further, in order to save and reduce the adjustment times, the server 104 also needs to set the transformer adjustment times constraint.

[0127]

[0128] wherein, K t is the transformer ratio, U is the standard voltage of the transformer j,t V is the voltage when the transformation ratio is K t σ is the standard deviation of the voltage i,t N is the number of gears, and σ is a binary variable indicating whether the gear changes σ is the difference between the square of the voltage standard of adjacent gears.

[0129] Further, the server 104 can also consider the group switching capacitor when establishing the model:

[0130]

[0131] Q is the reactive power of the group switching capacitor Q is the unit capacitor group power N is the number of switching groups at time t k,t N is the number of switching groups at the next time k,t+1 N is the maximum number of switching groups k,max N is the maximum number of switching groups N is a 0-1 variable indicating whether the number of switching groups changes, and is 1 when it changes, otherwise it is 0.

[0132] Further, the server 104 can also consider the static reactive power compensator when establishing the model:

[0133] Q is the reactive power of the static reactive power compensator SVC,min Q is the upper limit of the output of the static reactive power compensator SVC,t Q is the lower limit of the output of the static reactive power compensator SVC,max

[0134] Q is the reactive power of the static reactive power compensator SVC,t Q and Q are the upper and lower limits of the output of the static reactive power compensator SVC,max SVC,min

[0135] Through this double-layer optimization structure, the server 104 can obtain a complete scheduling scheme that considers both pre-disaster prevention and disaster response. This scheme has strong practicality and robustness, and can effectively respond to power distribution network failures under extreme ice and snow weather.

[0136] The upper model is a two-stage robust optimization model, which can be solved by the kkt function built-in yalmip. The main problem and the sub-problem of the robust optimization are iterated alternately until convergence, and the optimal solution of the upper model can be obtained. The upper model can be constructed by the above constraint conditions:

[0137]

[0138] C is the mobile energy storage pre-layout cost ME ​​​is a binary variable, and the value of 1 indicates that the mobile energy storage is connected to the node, and the value of 0 indicates that the mobile energy storage is disconnected from the node; w i is the unit load reduction cost of node i; P i Lshd is the load reduction active power of node i. The lower layer optimization objective of the double-layer optimization model is to minimize the load loss cost, which is constructed through the above constraint conditions:

[0139]

[0140] For example, in practical applications, if a similar failure to scenario 7 occurs (lines L5, L19, and L29 fail at the same time), the system operator can directly execute the pre-calculated dispatching scheme, including dispatching mobile energy storage according to the established route, starting the corresponding diesel generator, adjusting the load, and the like, so as to quickly and effectively restore power supply to important loads.

[0141] In the above power distribution system dispatching method applied to extreme weather, weather data and power distribution line parameters are obtained; wherein the power distribution line parameters include resistance strength parameters of multiple power distribution lines, which can comprehensively grasp the physical characteristics of the power distribution network and the current meteorological conditions faced, determine the failure probability of the multiple power distribution lines according to the weather data and the resistance strength parameters, can quantitatively evaluate the potential impact of extreme snow weather on each power distribution line, simulate the failure scenarios multiple times according to the failure probability of each power distribution line, and obtain multiple line failure scenario combinations; wherein each line failure scenario combination includes a predicted value of the failure of each power distribution line, and through multiple simulations, a statistically representative failure scenario set can be generated. The multiple line failure scenario combinations are input into the power distribution dispatching optimization model trained according to the failure prevention schemes and the failure handling schemes corresponding to multiple historical line failure scenario combinations, to obtain a power distribution dispatching scheme. Through the probability model and the multiple scenario simulation, various possible failure combinations and their occurrence probabilities are comprehensively considered, and the power distribution dispatching optimization model considering the failure prevention schemes and the failure handling schemes is used to determine the dispatching scheme, thereby improving the ability of the system to cope with extreme snow weather and improving the power supply safety of the power distribution system under extreme weather.

[0142] In one exemplary embodiment, as shown in FIG. 3, after step S206, steps S302 to S304 can be further included. Wherein: Figure 4

[0143] Step S302, clustering processing is performed on the multiple line failure scenario combinations to obtain multiple clustering clusters.

[0144] ​Exemplarily, the server 104 can reduce the large number of fault scenarios generated in the above step S206, select representative scenarios for subsequent optimization calculation, so as to reduce the calculation complexity while maintaining the quality and reliability of the solution. For example, the K-modes clustering algorithm is used for scenario reduction. The K-modes algorithm is a variant of the K-means algorithm and can be used to solve the clustering problem of categorical data. Unlike numerical data, categorical data is generally discrete, such as the classification of colors, product brands, and occupations. For lines, there are only two states of fault and non-fault, and the K-modes algorithm can be used for fault scenario reduction.

[0145] Further, the server 104 can obtain clustering parameters, and select a preset number of line fault scenario combinations as initial clustering centers according to the clustering parameters to form a plurality of initial clustering clusters; for each line fault scenario combination, calculate the distance value of the line fault scenario combination and each initial clustering center, and assign the line fault scenario combination to the initial clustering cluster corresponding to the smallest distance value; in the case that all line fault scenario combinations have been assigned to each initial clustering cluster, update each initial clustering center, and repeat the step of calculating the distance value of the line fault scenario combination and each initial clustering center for each line fault scenario combination until a preset stop condition is met, to obtain a plurality of clustering clusters. Exemplarily, the server 104 can obtain clustering parameters such as clustering number K (i.e. the number of fault scenarios expected to be obtained), dissimilarity measure definition (distance value), convergence threshold, etc. Then, the server 104 can select a preset number (K) of line fault scenario combinations as initial clustering centers according to the clustering parameters to form K initial clustering clusters. The selection of the initial clustering centers can be random or can use a specific strategy, such as selecting scenarios with relatively uniform fault number distribution.

[0146] Next, the server 104 can calculate the distance value of the line fault scenario combination and each initial clustering center for each line fault scenario combination, and assign the line fault scenario combination to the initial clustering cluster corresponding to the smallest distance value. The distance measure here can use the number of lines with different states in the two scenarios. For example, in scenario A, lines L1, L3, and L5 are faulty, and in scenario B, lines L2, L3, and L4 are faulty, then their distance value is 4 (the states of lines L1, L2, L4, and L5 are different).

[0147] In the case that all line fault scenario combinations have been assigned to the initial clustering clusters, the server 104 can update the initial clustering centers. The rule of updating the clustering centers can be that for each category attribute (i.e. whether each line is faulty), the value with the highest frequency of occurrence within the cluster is selected as the attribute value corresponding to the new clustering center. For example, if in a certain clustering cluster, line L1 is faulty in 60% of the scenarios, then in the updated clustering center, the status of line L1 can be faulty. The server 104 can repeat the step of calculating the distance value of each line fault scenario combination from the initial clustering centers for each line fault scenario combination until a preset stopping condition is met, obtaining a plurality of clustering clusters. The preset stopping condition can be that the number of iterations reaches an upper limit, or the change in the clustering center is less than a certain threshold.

[0148] Finally, the server 104 can obtain K clustering clusters, each cluster containing a group of similar fault scenarios. The server 104 can select a representative scenario from each cluster. For example, the actual scenario closest to the clustering center can be formed into a reduced fault scenario set. For example, if the original scenario set contains 1000 scenarios, it can be reduced to 10 representative scenarios through clustering.

[0149] Exemplarily, assume that the original set of scenarios contains 6 scenarios (in a real application, there can be hundreds or thousands of scenarios), each describing the status of 5 lines (0 means failure, 1 means normal): Scenario 1: [0, 1, 1, 1, 1] (only line L1 is failed); Scenario 2: [1, 0, 1, 1, 1] (only line L2 is failed); Scenario 3: [0, 0, 1, 1, 1] (lines L1 and L2 are failed); Scenario 4: [1, 1, 0, 0, 1] (lines L3 and L4 are failed); Scenario 5: [0, 1, 0, 1, 0] (lines L1, L3 and L5 are failed); Scenario 6: [1, 0, 0, 1, 0] (lines L2, L3 and L5 are failed). Set K = 2, randomly select Scenario 1 and Scenario 4 as the initial cluster centers. Calculate the distance values of each scenario to the two centers: Scenario 2 to center 1: 2, to center 2: 3, assigned to cluster 1; Scenario 3 to center 1: 1, to center 2: 4, assigned to cluster 1; Scenario 5 to center 1: 2, to center 2: 3, assigned to cluster 1; Scenario 6 to center 1: 3, to center 2: 2, assigned to cluster 2; update the cluster centers: cluster 1 contains scenarios 1, 2, 3, 5, the new center is [0, 1, 1, 1, 1] (the most common status of line L1 is failure, and the most common status of other lines is normal); cluster 2 contains scenarios 4, 6, the new center is [1, 1, 0, 1, 0] (the most common status of lines L3 and L5 is failure, and the most common status of other lines is normal). Continue the iteration until the cluster centers no longer change or the maximum number of iterations is reached. Finally, the server 104 can select Scenario 1 and Scenario 6 as the representative scenarios of the two clusters, forming the reduced set of scenarios.

[0150] By the K-modes clustering algorithm, the server 104 can successfully reduce the original large number of failure scenarios to a small number of representative scenarios, both reducing the complexity of subsequent optimization calculations and preserving the main features of the original set of scenarios.

[0151] Step S304, for each cluster, select the line failure scenario combination with the smallest distance value to the cluster center as the target failure scenario combination.

[0152] Exemplarily, the server 104 can calculate the distance values of all scenarios in each cluster to the cluster center. For each failure scenario combination, the server 104 can calculate the distance data of it to the cluster center to which it belongs, i.e. the number of lines with different states in the two scenarios. For example, if a scenario has lines L1, L3, L5 failed, and the cluster center has lines L1, L2, L5 failed, then their distance data is 2 (lines L2 and L3 have different states).

[0153] Subsequently, the server 104 can select, in each cluster, the line fault scenario combination with the smallest distance value to the cluster center as the target fault scenario combination. If there are multiple scenarios with the same and smallest distance to the cluster center, the server 104 can select any one of them as a representative, or select based on additional criteria (such as the number of fault lines, the importance of the fault lines, etc.). This selection method ensures that the target fault scenario combination can best represent the characteristics of the cluster it belongs to.

[0154] Further, step S208 can include step S306: inputting the plurality of target fault scenario combinations into the power distribution scheduling optimization model to obtain a power distribution scheduling scheme.

[0155] As described previously, the power distribution scheduling optimization model adopts a double-layer optimization structure, including a fault prevention sub-model (upper layer) and a fault handling sub-model (lower layer). Illustratively, the server 104 can input the target fault scenario combination into the fault prevention sub-model to obtain power distribution network topology and energy storage pre-layout data; and then input the power distribution network topology and energy storage pre-layout data into the fault handling sub-model to obtain detailed scheduling schemes under each target fault scenario.

[0156] In another exemplary embodiment, an IEEE33 node power distribution network is used for example simulation, and the line lengths are as shown in Table 1. Figure 5 As shown in Table 1, the line lengths are as shown in Table 1. Photovoltaic is connected at nodes 4, 17, 18, 20, and 22, diesel generators and static var compensators are connected at nodes 6, 16, and 32, and group switching capacitors are connected at nodes 6 and 16. A coordinate system is established with node 1 as the origin, and the ice storm center will move from (-24, -18) to (58, 22) at a speed of 4 km / h, with a maximum influence radius of 15 km. According to the line icing thickness calculation formula and line vulnerability, the line fault probability can be calculated, as shown in Table 1, where L5, etc. represent the line numbers that may fail.

[0157] Table 1 Line fault probability

[0158]

[0159] A large number of fault scenarios are generated using Monte Carlo simulation and are reduced by K-modes clustering. The reduced fault scenarios are shown in Table 2.

[0160] Table 2 Fault scenarios

[0161]

[0162]

[0163] The double-layer optimization method is used for scheduling for the fault scenarios 1-10, the number of fault lines in scenarios 1-6 is not more than 2, no island is formed after reconstruction, and the important load power demand can be met by the two mobile energy storages. Figure 6 The upper-layer optimization result of scenarios 7-10 is as shown in Figure 7 The lower-layer optimization result is as shown in Figure 8 As shown in

[0164] It should be understood that, although each step in the flowchart involved in each embodiment as described above is displayed in sequence according to the arrow, these steps are not necessarily executed in sequence according to the arrow. Unless explicitly stated herein, the execution of these steps is not strictly limited in sequence, and these steps can be executed in other sequences. Moreover, at least part of the steps in the flowchart involved in each embodiment as described above can include multiple steps or stages, which are not necessarily executed at the same time, but can be executed at different times, and the execution sequence of these steps or stages is not necessarily sequential, but can be alternately executed with at least part of other steps or steps or stages in other steps. It can be understood that each step in different embodiments can be freely combined as needed, and various non-contradictory schemes formed by the combination are within the scope of protection of the present application.

[0165] Based on the same inventive concept, the embodiments of the present application also provide a power distribution system scheduling device for extreme weather applied to the power distribution system scheduling method for extreme weather as described above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme described in the above method, so the specific limitations in one or more power distribution system scheduling device embodiments for extreme weather provided below can refer to the limitations of the power distribution system scheduling method for extreme weather described above, and will not be repeated here.

[0166] In an exemplary embodiment, as Figure 9As shown, a power distribution system scheduling device applied to extreme weather is provided, comprising: a data acquisition module 402, a fault analysis module 404, a fault prediction module 406 and a scheduling module 408, wherein: the data acquisition module 402 is configured to acquire weather data and power distribution line parameters; wherein the power distribution line parameters include the resistance strength parameters of multiple power distribution lines; the fault analysis module 404 is configured to determine the fault probability of the multiple power distribution lines according to the weather data and the resistance strength parameters; the fault prediction module 406 is configured to simulate the fault scene multiple times according to the fault probability of each power distribution line to obtain multiple line fault scene combinations; wherein each line fault scene combination includes the predicted value of the fault of each power distribution line; and the scheduling module 408 is configured to input the multiple line fault scene combinations into a power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is trained according to the fault prevention scheme and the fault handling scheme corresponding to multiple historical line fault scene combinations.

[0167] In one embodiment, the power distribution scheduling optimization model includes a fault prevention sub-model and a fault handling sub-model; and the scheduling module 408 is specifically configured to: input the multiple line fault scene combinations into the fault prevention sub-model to obtain power distribution network topology and energy storage pre-layout data; and input the power distribution network topology and energy storage pre-layout data into the fault handling sub-model to obtain the power distribution scheduling scheme under each line fault scene.

[0168] In one embodiment, the device further comprises a clustering processing module configured to perform clustering processing on the multiple line fault scene combinations to obtain multiple clustering clusters; for each clustering cluster, select a line fault scene combination with the smallest distance value from the clustering center as a target fault scene combination; and the scheduling module 408 is specifically configured to: input the multiple target fault scene combinations into the power distribution scheduling optimization model to obtain the power distribution scheduling scheme.

[0169] In one embodiment, the clustering processing module is specifically configured to: acquire clustering parameters, and select a preset number of line fault scene combinations as initial clustering centers according to the clustering parameters to form multiple initial clustering clusters; for each line fault scene combination, calculate the distance value between the line fault scene combination and each initial clustering center, and assign the line fault scene combination to the initial clustering cluster corresponding to the smallest distance value; in the case that all line fault scene combinations have been assigned to the initial clustering clusters, update each initial clustering center, and repeat the step of calculating the distance value between the line fault scene combination and each initial clustering center for each line fault scene combination until a preset stop condition is met to obtain the multiple clustering clusters.

[0170] In one of the embodiments, the fault prediction module 406 is specifically configured to generate, in each scenario simulation, a random number within a numerical range of the fault probability for each power distribution line; determine that the simulation result of the power distribution line in the scenario simulation is a fault in a case where the fault probability is greater than or equal to the random number; determine that the simulation result of the power distribution line in the scenario simulation is no fault in a case where the fault probability is less than the random number; and combine the simulation results of the power distribution lines in the scenario simulation to obtain the line fault scenario combination.

[0171] In one of the embodiments, the fault analysis module 404 is specifically configured to calculate an icing thickness prediction value according to the weather data; and input the icing thickness prediction value and the resistance strength parameter of each power distribution line into the line fault prediction model to obtain the fault probabilities of the plurality of power distribution lines.

[0172] The above-mentioned various modules in the power distribution system scheduling device applied to extreme weather can be realized by software, hardware, and combinations thereof, in whole or in part. The above-mentioned various modules can be embedded in or independent of a processor in a computer device in a hardware form, or can be stored in a memory in a computer device in a software form, so as to be called and executed by a processor to perform the operations corresponding to the above-mentioned various modules.

[0173] In one exemplary embodiment, a computer device, which can be a server, is provided, and an internal structure diagram of the computer device can be as shown in Figure 10 The computer device includes a processor, a memory, an input / output interface (I / O), and a communication interface. The processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. The processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The database of the computer device is configured to store historical and current weather data and power distribution line parameters and the like. The communication interface of the computer device is configured to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a power distribution system scheduling method applied to extreme weather.

[0174] Those skilled in the art can understand that Figure 10 The structure shown in the above-mentioned figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. Specifically, the computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0175] In an embodiment, a computer device is also provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0176] In an embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to implement the steps in the above method embodiments.

[0177] In an embodiment, a computer program product is provided, which comprises a computer program, and the computer program is executed by a processor to implement the steps in the above method embodiments.

[0178] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the processes of the above-mentioned embodiment methods. Any reference to memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. The non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. The volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a block chain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., without being limited thereto.

[0179] The technical features of the above embodiments can be combined in any manner. To make the description concise, not all possible combinations of the technical features in the above embodiments are described, but as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present application.

[0180] The above-described embodiments are merely illustrative of several embodiments of the present application, and the description is relatively specific and detailed, but should not be understood as a limitation on the scope of the patent. It should be noted that for those skilled in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are all within the scope of the present application. Therefore, the scope of protection of the present application should be subject to the appended claims.

Claims

1. A power distribution system dispatching method for extreme weather, characterized by, The method comprises: obtaining weather data and power distribution line parameters; wherein the power distribution line parameters comprise resistance strength parameters of multiple power distribution lines; determining failure probabilities of the multiple power distribution lines according to the weather data and the resistance strength parameters; simulating failure scenarios multiple times according to the failure probabilities of the power distribution lines to obtain multiple line failure scenario combinations; wherein each line failure scenario combination comprises predicted values of failures of the power distribution lines; inputting the multiple line failure scenario combinations into a power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is trained according to historical line failure scenario combinations and corresponding failure prevention schemes and failure handling schemes.

2. The method of claim 1, wherein, The power distribution scheduling optimization model comprises a failure prevention sub-model and a failure handling sub-model. The inputting of the multiple line failure scenario combinations into the power distribution scheduling optimization model to obtain the power distribution scheduling scheme comprises: inputting the multiple line failure scenario combinations into the failure prevention sub-model to obtain power distribution network topology and energy storage pre-layout data; inputting the power distribution network topology and the energy storage pre-layout data into the failure handling sub-model to obtain power distribution scheduling schemes under each line failure scenario.

3. The method of claim 1, wherein, After the simulating of the failure scenarios multiple times according to the failure probabilities of the power distribution lines to obtain the multiple line failure scenario combinations, the method further comprises: performing clustering processing on the multiple line failure scenario combinations to obtain multiple clustering clusters; for each clustering cluster, selecting a line failure scenario combination with the smallest distance value from a clustering center as a target failure scenario combination; The inputting of the multiple line failure scenario combinations into the power distribution scheduling optimization model to obtain the power distribution scheduling scheme comprises: inputting the multiple target failure scenario combinations into the power distribution scheduling optimization model to obtain the power distribution scheduling scheme.

4. The method of claim 3, wherein, The clustering processing on the multiple line failure scenario combinations to obtain the multiple clustering clusters comprises: obtaining clustering parameters and selecting a preset number of line failure scenario combinations as initial clustering centers according to the clustering parameters to form multiple initial clustering clusters; for each line failure scenario combination, calculating distance values of the line failure scenario combination from the initial clustering centers and assigning the line failure scenario combination to an initial clustering cluster corresponding to the smallest distance value; under the condition that all the line failure scenario combinations have been assigned to the initial clustering clusters, updating the initial clustering centers, repeating the step of calculating the distance values of the line failure scenario combination from the initial clustering centers for each line failure scenario combination, and obtaining the multiple clustering clusters until a preset stop condition is met.

5. The method according to any one of claims 1 to 4, characterized in that, In each simulation of a scenario, for each power distribution line, a random number in a numerical range of the failure probability is generated. ​ In a case where the failure probability is greater than or equal to the random number, it is determined that the simulation result of the power distribution line in the scenario simulation is a failure; In a case where the failure probability is less than the random number, it is determined that the simulation result of the power distribution line in the scenario simulation is no failure; The simulation results of each of the power distribution lines in the scenario simulation are combined to obtain a line failure scenario combination.

6. The method according to any one of claims 1 to 4, characterized in that, The method further includes: According to the weather data, an ice thickness prediction value is calculated; The ice thickness prediction value and the resistance strength parameter of each of the power distribution lines are input into a line failure prediction model to obtain the failure probability of the plurality of power distribution lines.

7. A power distribution system dispatching device for extreme weather applications, characterized by, The apparatus includes: A data acquisition module configured to acquire weather data and power distribution line parameters; wherein the power distribution line parameters include resistance strength parameters of a plurality of power distribution lines; A failure analysis module configured to determine failure probabilities of the plurality of power distribution lines according to the weather data and the resistance strength parameters; A failure prediction module configured to simulate a failure scenario multiple times according to the failure probability of each of the power distribution lines to obtain a plurality of line failure scenario combinations; wherein each of the line failure scenario combinations includes a prediction value of a failure of each of the power distribution lines; A scheduling module configured to input the plurality of line failure scenario combinations into a power distribution scheduling optimization model to obtain a power distribution scheduling scheme; wherein the power distribution scheduling optimization model is trained according to a plurality of historical line failure scenario combinations corresponding to failure prevention schemes and failure handling schemes.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to implement the steps of the method of any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method of any one of claims 1 to 6.

Images