Multi-resource collaborative power distribution network scheduling method, system, equipment, medium and product

Abstract

The invention relates to the technical field of power distribution networks, and discloses a multi-resource collaborative power distribution network scheduling method, system, device, medium and product, and the method comprises the steps: determining a typical fault scene of a power distribution network through collecting the operation fault data of each element of the power distribution network in a typhoon weather scene; and minimization of load loss of the power distribution network, minimization of total scheduling cost and minimization of key load recovery time are taken as multiple optimization objectives, a load loss rate, economic cost and key load recovery time are taken into consideration at the same time, economy-toughness-aging multi-dimensional balance is taken into consideration, and the power distribution network and the EV charging station are optimized according to the balance operation state of the power distribution network and the EV charging station. According to the method, the constraint condition of multiple optimization targets is determined, so that the operation reliability of the power distribution network and the EV charging station is ensured, a power distribution network scheduling optimization scheme is obtained by constructing a power distribution network scheduling optimization model and performing optimization solution, so that the real-time scheduling of multi-element flexible resources is realized, and the power distribution network scheduling reliability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution networks, and particularly to a distribution network scheduling method, system, device, medium and product for collaborative multi-source resources. Background Art

[0002] In recent years, the frequent occurrence of extreme weather events (such as typhoons, heavy rains, etc.) has put forward higher requirements for the stable operation of distribution networks. At the same time, the continuous increase in the penetration rate of renewable energy generation (such as distributed photovoltaics, wind power, etc.) has further increased the uncertainty and complexity of power grid operation. Extreme meteorological conditions such as typhoons not only pose threats to traditional transmission lines and substations, but also seriously affect the normal operation of distributed energy equipment, thus exacerbating the vulnerability of distribution networks in disaster scenarios.

[0003] Although traditional numerical forecasting, statistical models and some machine learning methods can reflect the macroscopic characteristics of disasters such as typhoons to a certain extent, they still have deficiencies in capturing the spatio-temporal characteristics of typhoons, predicting disaster evolution and evaluating the impact of disasters on the failures of power grid components (such as transmission lines, substations, photovoltaic systems). At the same time, with the popularization of flexible resources such as distributed energy, energy storage devices, mobile energy storage and electric vehicles, how to fully coordinate these multi-source resources in disaster scheduling to achieve a balance among economy, resilience and timeliness has become an important research direction for current power grid scheduling and restoration.

[0004] In existing distribution network scheduling, it is difficult to achieve efficient distributed coordination and real-time scheduling for real-time scheduling of multi-source flexible resources (such as electric vehicles (EVs), energy storage, mobile energy storage), and the reliability of scheduling is poor. Summary of the Invention

[0005] In view of this, the present invention provides a distribution network scheduling method, system, device, medium and product for collaborative multi-source resources, which solves the technical problems that it is difficult to achieve efficient distributed coordination and real-time scheduling for real-time scheduling of multi-source flexible resources, and the reliability of scheduling is poor.

[0006] The first aspect of the present invention provides a distribution network scheduling method for collaborative multi-source resources, including:

[0007] Collect the operation failure data of each component of the distribution network in the typhoon weather scenario, and determine the typical failure scenarios of the distribution network according to the operation failure data;

[0008] Take the minimization of load loss, the minimization of total scheduling cost and the shortest restoration time of critical loads of the distribution network as multiple optimization objectives;

[0009] Determine the constraint conditions of the multi-optimization objectives according to the balanced operation status of the distribution network and the EV charging station;

[0010] Construct a distribution network scheduling optimization model according to the multi-optimization objectives and the constraint conditions;

[0011] Perform optimization solution on the distribution network scheduling optimization model according to the typical fault scenarios, and obtain a distribution network scheduling optimization plan under the typical fault scenarios.

[0012] Preferably, the operation fault data includes line fault probability and photovoltaic unit fault probability;

[0013] Collect the operation fault data of each component of the distribution network in the typhoon weather scenario, and determine the typical fault scenarios of the distribution network according to the operation fault data, including:

[0014] Predict the typhoon predicted wind speed at the current prediction moment through a pre-trained typhoon time-varying prediction model according to the historical typhoon wind speed;

[0015] Construct a line fault probability model and a photovoltaic unit fault probability model respectively, and determine the line fault probability and the photovoltaic unit fault probability according to the line fault probability model and the photovoltaic unit fault probability model, in combination with the typhoon predicted wind speed;

[0016] Generate multiple fault scenarios by Monte Carlo sampling according to the line fault probability and the photovoltaic unit fault probability;

[0017] Select the fault scenario corresponding to the maximum entropy value of the line fault probability and the photovoltaic unit fault probability from multiple fault scenarios, and determine the typical fault scenario of the distribution network.

[0018] Preferably, the method further includes:

[0019] Collect typhoon historical sample time series data, and extract spatio-temporal features from the typhoon historical sample time series data through a convolutional neural network;

[0020] Input the spatio-temporal features into a long short-term memory network to process the time series relationship in the spatio-temporal features, and obtain a typhoon time-varying prediction model.

[0021] Preferably, the objective function corresponding to the multi-optimization objectives is:

[0022]

[0023] In the formula, is the weighted load loss rate, is the total scheduling cost, is the critical load restoration time; , , are all weight coefficients;

[0024] Among them,

[0025] In the formula, is the active power loss of node n at time t, is the active power demand of node n at time t, is the load importance weight of node n, is the total number of time periods in the scheduling cycle, is the total number of distribution network nodes;

[0026]

[0027] In the formula, is the EV scheduling cost at time; is the energy storage scheduling cost at time; is the PV unit scheduling cost at time; is the charging station incentive cost;

[0028] Among them,

[0029] In the formula, is the unit discharge loss cost of the battery of EV charging station q, is the set of EV charging stations at node , is the active power output by EV charging station q at node at time ;

[0030]

[0031] In the formula, is the unit scheduling cost of MESS; is the active power output by MESS at node at time , is the unit distance scheduling cost, is the scheduling distance; is the number of users participating in scheduling; m is the index of users participating in scheduling;

[0032]

[0033] In the formula, is the unit output operation and maintenance cost of the PV unit; is at time at node The active power output at the PV

[0034] Among them,

[0035]

[0036] In the formula, is the dynamic incentive electricity price of node , is the basic incentive electricity price; is the incentive electricity price adjustment coefficient; is the active power output of all EV charging stations at node at time

[0037]

[0038] In the formula, is the unit time cost of the critical node delay recovery; is the critical load node set; is the duration from the power outage to the power restoration of node n.

[0039] Preferably, the constraint conditions include power grid operation constraints and EV charging station scheduling constraints; among them, the power grid operation constraints include topological constraints, node power balance constraints, node voltage relaxation constraints, line transmission power limit constraints, node voltage limit constraints, node load shedding limit constraints, and PV operation limit constraints;

[0040] The EV charging station scheduling constraints include road network-power grid coupling constraints, EV power quantity limit constraints participating in scheduling, user discharge response rate constraints under the incentive electricity price, balance constraints between the user discharge response rate and the EV charging station output under the incentive electricity price, EV charge and discharge power limit constraints, EV power quantity limit constraints, and MESS scheduling limit constraints;

[0041] Among them, the road network-power grid coupling constraint is used to constrain that the EV arrives at the specified location in time for charging and discharging during the scheduling process;

[0042] The EV power quantity limit constraint participating in scheduling is used to ensure that the EV power quantity is not lower than the preset threshold at the end of the scheduling;

[0043] The user discharge response rate constraint under the incentive electricity price is used to constrain and determine the user discharge response rate under different incentive electricity prices according to the sensitivity of the user to the incentive electricity price;

[0044] The balance constraint between the user discharge response rate and the EV charging station output under the incentive electricity price is used to constrain and ensure that the output of the EV charging station matches the user's discharge demand.

[0045] Preferably, according to the typical fault scenario, optimizing and solving the distribution network scheduling optimization model to obtain a distribution network scheduling optimization plan under the typical fault scenario, including:

[0046] Randomly generate multiple distribution network scheduling candidate plans that meet the constraint conditions, and initialize the population using the multiple distribution network scheduling candidate plans;

[0047] Perform non-dominated sorting on the individuals in the initialized population, and calculate the crowding distance;

[0048] Determine the fitness value of each individual according to the objective function corresponding to the multiple optimization objectives;

[0049] According to the fitness value and the crowding distance, select individuals with a fitness value greater than the preset fitness threshold and a crowding distance greater than the preset crowding threshold to form a new population;

[0050] Perform crossover and mutation operations on the new population to generate new offspring individuals;

[0051] Merge the new offspring individuals with the parent individuals to form a new population, and repeat performing non-dominated sorting on the individuals in the initialized population and calculating the crowding distance until the iteration termination condition is met;

[0052] Select the optimal individual from the final population as the optimal solution of the distribution network scheduling optimization model to obtain a distribution network scheduling optimization plan under the typical fault scenario.

[0053] In a second aspect, the present invention provides a distribution network scheduling system for multi-source resource collaboration, including:

[0054] A fault scenario determination module, configured to collect the operation fault data of each component of the distribution network in a typhoon weather scenario, and determine the typical fault scenario of the distribution network according to the operation fault data;

[0055] An optimization objective determination module, with the minimization of load loss of the distribution network, the minimization of the total scheduling cost, and the shortest restoration time of critical loads as multiple optimization objectives;

[0056] A constraint determination module, configured to determine the constraint conditions of the multiple optimization objectives according to the balanced operation state of the distribution network and the EV charging station;

[0057] A model construction module, configured to construct a distribution network scheduling optimization model according to the multiple optimization objectives and the constraint conditions;

[0058] A scheduling optimization module, configured to optimize and solve the distribution network scheduling optimization model according to the typical fault scenarios, so as to obtain a distribution network scheduling optimization scheme under the typical fault scenarios.

[0059] In a third aspect, the present invention provides an electronic device, which includes a memory and a processor. A computer program is stored in the memory. When the computer program is executed by the processor, the processor is caused to execute the steps of the distribution network scheduling method for multi-resource collaboration as described in the first aspect.

[0060] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the steps of the distribution network scheduling method for multi-resource collaboration as described in the first aspect are implemented.

[0061] In a fifth aspect, the present invention provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the steps of the distribution network scheduling method for multi-resource collaboration as described in the first aspect.

[0062] As can be seen from the above technical solutions, the present invention determines the typical fault scenarios of the distribution network by collecting the operation fault data of each component of the distribution network in typhoon weather scenarios, and takes the minimization of the load loss of the distribution network, the minimization of the total scheduling cost, and the shortest critical load recovery time as multiple optimization objectives, simultaneously considering the load loss rate, economic cost, and critical load recovery time, taking into account the multi-dimensional balance of economy-resilience-timeliness. By determining the constraint conditions of the multiple optimization objectives according to the balanced operation states of the distribution network and EV charging stations, the operation reliability of the distribution network and EV charging stations is ensured. Through the multiple optimization objectives and constraint conditions, a distribution network scheduling optimization model is constructed and optimized to solve, and a distribution network scheduling optimization scheme is obtained, thereby realizing the real-time scheduling of multiple flexible resources, and also realizing efficient distributed coordination and real-time scheduling, improving the reliability of distribution network scheduling. Description of the Drawings

[0063] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0064] Figure 1 It is an application environment diagram of a distribution network scheduling method for multi-resource collaboration provided by an embodiment of the present invention;

[0065] Figure 2 This is a flowchart of a multi - resource collaborative distribution network scheduling method provided by an embodiment of the present invention;

[0066] Figure 3 This is a schematic structural diagram of a multi - resource collaborative distribution network scheduling system provided by an embodiment of the present invention;

[0067] Figure 4 This is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. Detailed implementation manners

[0068] In order to enable those skilled in the art to better understand the solution of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts shall fall within the protection scope of the present invention.

[0069] The multi - resource collaborative distribution network scheduling method provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 101 communicates with the server 102 through a network. The data storage system can store the data that the server 102 needs to process. The data storage system can be integrated on the server 102, or placed in the cloud or other network servers. The terminal 101 or the server 102 collects the operation failure data of each component of the distribution network in the typhoon weather scenario, determines the typical failure scenarios of the distribution network according to the operation failure data; takes the minimization of the load loss of the distribution network, the minimization of the total scheduling cost, and the shortest restoration time of critical loads as multiple optimization objectives; determines the constraint conditions of the multiple optimization objectives according to the balanced operation states of the distribution network and the EV charging stations; constructs a distribution network scheduling optimization model according to the multiple optimization objectives and the constraint conditions; and performs optimization solution on the distribution network scheduling optimization model according to the typical failure scenarios to obtain a distribution network scheduling optimization plan under the typical failure scenarios.

[0070] The terminal 101 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, etc.

[0071] The server 102 can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0072] As Figure 2 shown, the embodiments of the present application provide a multi - resource collaborative distribution network scheduling method, and this method is applied toFigure 1 Taking the terminal 101 or the server 102 in

[0073] Step S1: Collect the operation fault data of each component of the distribution network in the typhoon weather scenario, and determine the typical fault scenario of the distribution network according to the operation fault data.

[0074] Among them, the components of the distribution network refer to various devices and components in the distribution network, including but not limited to transformers, switchgear, lines, capacitors, reactors, and photovoltaic (PV) etc. By collecting the operation fault data of these components in typhoon weather, the operation status and possible fault conditions of the distribution network under extreme weather conditions can be comprehensively understood. The operation fault data includes but not limited to the fault probability of the components, etc., and the typical fault scenario that is most likely to occur in the distribution network in typhoon weather can be determined.

[0075] The typical fault scenario refers to the most representative and most likely fault condition analyzed according to the operation fault data of each component of the distribution network in typhoon weather.

[0076] Step S2: Take the minimization of the load loss of the distribution network, the minimization of the total dispatching cost, and the shortest restoration time of critical loads as multiple optimization objectives.

[0077] After determining the typical fault scenario in the embodiment of the present application, it is necessary to determine the objectives of the optimal dispatching. In this embodiment, the minimization of the load loss of the distribution network, the minimization of the total dispatching cost, and the shortest restoration time of critical loads are taken as multiple optimization objectives, so as to realize the dispatching optimization of the distribution network.

[0078] Among them, the minimization of the load loss means that on the premise of ensuring the safe and stable operation of the power grid, the power outage time and outage range of the load are reduced as much as possible to improve the power supply reliability; the minimization of the total dispatching cost means that on the premise of meeting the load demand and the power grid safety constraints, the cost generated in the dispatching process is reduced as much as possible, including EV dispatching cost, energy storage dispatching cost, photovoltaic unit dispatching cost, and charging station incentive cost, etc.; the shortest restoration time of critical loads means that when a fault occurs, the power supply of critical loads is restored as soon as possible to reduce the impact on society and economy.

[0079] Specifically, the objective function corresponding to the multiple optimization objectives is:

[0080]

[0081] In the formula, is the weighted load loss rate, is the total dispatching cost, is the critical load restoration time; , , are all weight coefficients; among them, .

[0082] Among them,

[0083] In the formula, is the active power loss of node n at time t, is the active power demand of node n at time t, is the load importance weight of node n, is the total number of time periods in the scheduling cycle, is the total number of distribution network nodes;

[0084]

[0085] In the formula, is the EV scheduling cost at time is the energy storage scheduling cost at time is the PV unit scheduling cost at time is the charging station incentive cost;

[0086] Among them,

[0087] In the formula, is the unit discharge loss cost of the battery of EV charging station q, is the set of EV charging stations at node , is the active power output by EV charging station q at node at time

[0088]

[0089] In the formula, is the unit scheduling cost of MESS; is the active power output by MESS at node at time is the unit distance scheduling cost, is the scheduling distance; is the number of users participating in scheduling; m is the index of users participating in scheduling;

[0090]

[0091] In the formula, is the unit output operation and maintenance cost of the PV unit; is the active power output by the PV unit at node at time The active power output at the PV

[0092] Among them,

[0093]

[0094] In the formula, is the dynamic incentive electricity price of node , is the basic incentive electricity price; is the incentive electricity price adjustment coefficient; among them, is dynamically adjusted according to the load loss, and is used to dynamically increase the incentive electricity price according to the load loss rate; is at time the active power output of all EV charging stations at node

[0095]

[0096] In the formula, is the unit time cost of the delayed restoration of the critical node; is the set of critical load nodes; is the duration from power outage to power restoration of node n.

[0097] Step S3: Determine the constraint conditions of multiple optimization objectives according to the balanced operation status of the distribution network and EV charging stations.

[0098] Among them, the constraint conditions of multiple optimization objectives consider the balanced operation status of the distribution network and electric vehicle (EV) charging stations, and ensure that the operation status of the distribution network after scheduling is safe and reliable. The balance constraint of the EV charging station is to ensure that during the scheduling process, the charging and discharging behaviors of the EV charging station can meet the needs of the power grid without affecting the normal use of EVs. These constraint conditions together constitute the boundary conditions of the distribution network scheduling optimization problem, ensuring that the optimization scheme is feasible and effective in practical applications.

[0099] Specifically, the constraint conditions include power grid operation constraints and EV charging station scheduling constraints; among them, the power grid operation constraints include topology constraints, node power balance constraints, node voltage relaxation constraints, line transmission power limit constraints, node voltage limit constraints, node load shedding limit constraints, and PV operation limit constraints.

[0100] Among them, the topological constraint means that during the dispatching process of the distribution network, the connection relationship between components remains unchanged to ensure the stability of the grid structure. The node power balance constraint means that the sum of the injected power and the outflow power at each node at any moment is zero to ensure the power balance of the grid. The node voltage relaxation constraint relaxes the node voltage to adapt to the voltage fluctuations during the dispatching process of the grid. The line transmission power limit constraint means that the transmission power of each line cannot exceed its maximum transmission capacity to prevent line overload. The node voltage limit constraint means that the voltage of each node must be within the specified range to ensure the stable operation of the grid. The node load shedding limit constraint restricts the possible load shedding situation of the node during the dispatching process to ensure the power supply of critical loads. The PV operation limit constraint means that the output of the PV unit during the dispatching process cannot exceed its maximum output to prevent the PV unit from being overloaded.

[0101] Among them, the topological constraint (radial operation) is as follows:

[0102]

[0103]

[0104]

[0105]

[0106] In the formula, is the branch state; and are the virtual power flow directions respectively; means that the actual power flow direction is the same as the default flow direction, means that the actual power flow direction is opposite to the default flow direction; is the set of all nodes in the island network; is the set of standby island power supply nodes; is the set of all branches connected to node i; is the set of all branches with node i as the child node.

[0107] The node power balance constraint is:

[0108]

[0109] In the formula, and are the downstream node set and the upstream node set of node respectively; and are respectively at time on line The active power and reactive power flowing at the moment; and are respectively the active power and reactive power flowing on line at the moment ; is the active power injected by the superior power grid at node at the moment ; is the reactive power output by the EV charging station at node at the moment ; is the reactive power output by the MESS at node at the moment ; is the reactive power output by the PV at node at the moment ; is the reactive power demand at node

[0110] ]>The node voltage relaxation constraint is:

[0111]

[0112] wherein, and are respectively the voltage values of node and node at the moment ; is the rated voltage value; is the fault state of line at the moment and are respectively the resistance and reactance on line

[0113] The line transmission power limit constraint and the node voltage limit constraint are respectively:

[0114]

[0115] wherein, is the maximum line transmission power; and are respectively the upper and lower voltage limit values of the node

[0116] The node load shedding constraint is as follows:

[0117]

[0118] The PV operation constraint is as follows:

[0119]

[0120] In the formula, and are respectively the maximum theoretical active power and reactive power of the PV at node The maximum values of active power and reactive power of the PV at node

[0121] The EV charging station scheduling constraint considers the EV travel characteristics, describes the EV scheduling behavior in the early stage of the disaster using the travel chain theory, obtains the EV spatio-temporal distribution state and the current electricity consumption, and ensures that during the scheduling process, the charging and discharging behaviors of the EV charging stations can meet the grid requirements without affecting the normal use of the EVs.

[0122] The EV charging station scheduling constraints include the road network - power grid coupling constraint, the EV power limit constraint for participating in the scheduling, the discharge response rate constraint of users under the incentive electricity price, the balance constraint between the discharge response rate of users under the incentive electricity price and the output of the EV charging station, the EV charge and discharge power limit constraint, the EV power limit constraint, and the MESS scheduling limit constraint;

[0123] Among them, the road network - power grid coupling constraint is used to ensure that the EVs can reach the designated locations for charging and discharging in a timely manner during the scheduling process;

[0124] The EV power limit constraint for participating in the scheduling is used to ensure that the EVs have no less than the preset threshold of electricity consumption at the end of the scheduling;

[0125] The discharge response rate constraint of users under the incentive electricity price is used to determine the discharge response rate of users under different incentive electricity prices according to the sensitivity of users to the incentive electricity price;

[0126] The balance constraint between the discharge response rate of users under the incentive electricity price and the output of the EV charging station is used to ensure that the output of the EV charging station matches the discharge demand of users.

[0127] The charging and discharging power limit constraints of EVs mean that during the charging and discharging processes of EVs, their power cannot exceed the maximum power that the charging station and the EV itself can withstand, in order to prevent equipment overload and damage. The state of charge limit constraints of EVs ensure that during the participation in the dispatching process, the state of charge of EVs remains within a certain range, neither too high nor too low, to ensure the normal use of EVs. The dispatching limit constraints for the Microgrid Energy Storage System (MESS) in the microgrid are for the dispatching of MESS, ensuring that the charging and discharging behaviors of MESS during the dispatching process can meet the grid's requirements, while ensuring the safe operation and energy storage efficiency of MESS. These constraint conditions together constitute the boundary conditions of the EV charging station dispatching optimization problem, ensuring that the optimization scheme is feasible and effective in practical applications.

[0128] Specifically, the road network - power grid coupling constraint is as follows:

[0129]

[0130]

[0131] In the formula, is the edge set of the road network and the power grid; is the set of coupling points of the road network and the power grid; is the set of road network nodes; is the set of power grid nodes; is the th node of the road network; is the th node of the power grid; represents the connection relationship between the road network and the power grid.

[0132] The construction process of the state of charge limit constraints of EVs participating in dispatching is as follows:

[0133] Travel chain modeling: The user's travel chain is described by state variables such as the starting point, ending point, path, and time:

[0134]

[0135] In the formula, is the set of state variables of a section of the travel chain; is the starting point of this section of the journey; is the ending point of this section of the journey; is the driving path; is the departure time; is the arrival time; is the driving duration; is the stopping duration.

[0136] Considering the For a certain section of the journey, it can be calculated as follows:

[0137]

[0138]

[0139]

[0140] Wherein, The Travel time for the section of the journey; The length of the section of the journey; The average speed of the section of the journey; The arrival time of the section of the journey; The departure time of the

[0141] Considering the travel willingness of EV users:

[0142]

[0143] Wherein, Is the travel willingness index, Is the travel willingness coefficient under disasters ( ); Is the total number of EVs in the region.

[0144] The travel time follows a normal distribution, and the real-time state of EVs can be modeled:

[0145]

[0146] Wherein, Is the set of real-time states of each EV at time; Is the travel time of each EV at time; Is the current battery level of each EV at time; Is the EV charging station closest to the current location.

[0147] EVs participating in power supply restoration need to meet the battery level constraint:

[0148]

[0149] Wherein, Is the rated discharge power of the EV charging station; Is the fault repair time. For example, It is set to 4h.

[0150] The construction process of the discharge response rate constraint of users under the incentive electricity price is as follows:

[0151] Based on the theory of consumer psychology, the discharge response of users shows a piecewise linear relationship with the incentive price:

[0152] The upper boundary function is:

[0153]

[0154] In the formula, is the upper limit of the discharge response rate of user m in the ideal scenario; is the incentive electricity price (yuan / kWh); is the lowest incentive electricity price threshold for user m to start responding in the ideal scenario; is the highest incentive electricity price threshold for user m to reach saturation in the ideal scenario; is the maximum response rate of user m in the ideal scenario when the incentive electricity price reaches

[0155] The lower boundary function is:

[0156]

[0157] In the formula, is the upper limit of the discharge response rate of user m in the conservative scenario; is the lowest incentive electricity price threshold for user m to start responding in the conservative scenario; is the highest incentive electricity price threshold for user m to reach saturation in the conservative scenario; is the maximum response rate of user m in the conservative scenario when the incentive electricity price reaches

[0158] Then the discharge response rate constraint of users under the incentive electricity price is:

[0159]

[0160] In the formula, is the actual discharge response rate of user m under the incentive electricity price, which follows a uniform distribution; is the symbol of the uniform distribution, indicating that the actual response rate fluctuates randomly between the upper and lower boundaries.

[0161] The balance constraint between the discharge response rate of users under the incentive electricity price and the output of EV charging stations is determined by the relationship between the response rate and the output of EV charging stations, and we get:

[0162] In the formula, is the number of EVs that meet the electricity demand at node ; ​​is the discharge power of a single EV.

[0163] The charging and discharging power limit constraints of the EV are:

[0164]

[0165] In the formula, and are respectively at time the charging and discharging powers of the th EV; and are respectively the upper limits of the charging and discharging powers of the EV; and are respectively the charging and discharging states of the

[0166] th

[0167]

[0168] In the formula, is at time the power of the th EV; and are respectively the upper and lower limits of the EV power;

[0169] The MESS scheduling limit constraint is:

[0170]

[0171]

[0172] In the formula, is the MESS state at node , indicates that there is an allocated MESS at this node, indicates that there is no allocated MESS at this node; and are respectively the upper limits of the active and reactive powers output by the MESS.

[0173] Step S4: Construct a distribution network scheduling optimization model according to multiple optimization objectives and constraint conditions.

[0174] Step S5: Optimize and solve the distribution network scheduling optimization model according to typical fault scenarios to obtain a distribution network scheduling optimization plan under typical fault scenarios.

[0175] It should be noted that in the embodiments of the present application, by collecting the operation failure data of each component of the distribution network in the typhoon weather scenario, the typical failure scenarios of the distribution network are determined, and with the minimization of the load loss of the distribution network, the total dispatching cost, and the shortest critical load recovery time as multiple optimization objectives, the load loss rate, economic cost, and critical load recovery time are simultaneously considered, taking into account the multi-dimensional balance of economy - resilience - timeliness. By determining the constraint conditions of the multiple optimization objectives according to the balanced operation status of the distribution network and EV charging stations, the operation reliability of the distribution network and EV charging stations is ensured. Through the multiple optimization objectives and constraint conditions, a distribution network dispatching optimization model is constructed and optimized to obtain a distribution network dispatching optimization plan, thereby realizing the real-time dispatching of multiple flexible resources and also achieving efficient distributed coordination and real-time dispatching, improving the reliability of distribution network dispatching.

[0176] In some embodiments, the operation failure data includes the line failure probability and the photovoltaic unit failure probability;

[0177] Collecting the operation failure data of each component of the distribution network in the typhoon weather scenario, and determining the typical failure scenarios of the distribution network according to the operation failure data, including:

[0178] Step S101: Predict through a pre-trained typhoon time-varying prediction model according to the historical typhoon wind speed to obtain the predicted typhoon wind speed at the current prediction moment.

[0179] Among them, the training process of the typhoon time-varying prediction model is as follows:

[0180] Step S1011: Collect historical typhoon sample time-series data, and extract spatio-temporal features from the historical typhoon sample time-series data through a convolutional neural network.

[0181] Among them, the spatio-temporal features include key information such as typhoon wind speed, typhoon movement path, typhoon intensity change, and typhoon influence range.

[0182] Among them, based on a multi-scale three-dimensional convolutional neural network (3D Convolutional Neural Network, 3DNN) to extract the spatio-temporal features of the historical typhoon sample time-series data to form the spatio-temporal correlation of atmospheric parameters during the typhoon formation process. The calculation process of the output value at position (x, y, z) of the jth feature map in the ith layer:

[0183]

[0184] In the formula, is the activation value of the jth feature map in the ith layer at position (x, y, z); k traverses all the input feature maps (channel dimension) of the layer, , where is the The number of feature maps of the layer; is a non-linear activation function; p, q, and r are the offsets of the three-dimensional convolution kernel in the longitude, latitude, and pressure layer dimensions respectively, , , , where is the convolution kernel size; is the weight of the j-th convolution kernel in the i-th layer on the k-th input feature map, located at (p, q, r); is the value of the k-th input feature map in the i-th layer at the position ; is the bias term of the j-th feature map in the i-th layer.

[0185] Parallel multi-scale convolution: Three different sizes of convolution kernels (3×3×3, 5×5×5, 7×7×7) are used to extract features in parallel, capturing local details, medium-scale structures, and global environmental fields respectively, to obtain:

[0186]

[0187] In the formula, is channel dimension concatenation, concatenating multiple feature maps in the channel dimension to fuse information of different scales; is the multi-scale feature map, which is obtained by concatenating the outputs of 3×3, 5×5, and 7×7 convolutions (64×3 = 192 channels); are the parameters of the 3×3×3 convolution kernel, with N input channels and 64 output channels; are the parameters of the 5×5×5 convolution kernel, with N input channels and 64 output channels; are the parameters of the 7×7×7 convolution kernel, with N input channels and 64 output channels.

[0188] Spatial attention weight generation:

[0189]

[0190] In the formula: is global average pooling, which compresses the feature map of each channel into a scalar; is the spatial attention weight, which is generated through global average pooling and a fully connected layer; , are the parameters of the fully connected layer for attention weight generation.

[0191] Among them,

[0192] In the formula, is the weighted feature map, that is, the spatio-temporal feature.

[0193] Step S1012: Input the spatio-temporal features into a long short-term memory network to process the time series relationship in the spatio-temporal features, and obtain a typhoon time-varying prediction model.

[0194] Among them, based on the spatio-temporal features extracted by 3DCNN, the formation and intensity of future typhoons are predicted. The output prediction formula is:

[0195]

[0196] In the formula, is the hidden state at the current time step, containing historical time series information; W is the weight matrix, b is the bias term, and they are learned through training; is the output mapping.

[0197] The state update formula of the long short-term memory network (Long Short-Term Memory, LSTM):

[0198]

[0199] In the formula, is the feature after attention enhancement; is the hidden state at the previous time step; is the memory cell at the previous time step, storing long-term dependence information; is the updated hidden state and memory cell, which are passed to the next time step.

[0200] The prediction model obtained by combining 3DCNN-LSTM is:

[0201]

[0202] Among them, is the typhoon time-varying feature at historical time , with a dimension of , is the resolution of the longitude and latitude grid, L is the number of pressure layers, and N is the number of atmospheric variables (wind speed, temperature, humidity).

[0203] Step S102: Construct a line fault probability model and a photovoltaic unit fault probability model respectively, and determine the line fault probability and the photovoltaic unit fault probability according to the line fault probability model and the photovoltaic unit fault probability model, combined with the predicted typhoon wind speed.

[0204] Among them, typhoon disasters mainly cause strong forces on the conductors and towers of transmission lines, resulting in the disconnection of the lines between towers. Model the line faults of the distribution network under extreme weather, divide the distribution network into a grid, divide it into equal-side-length grids according to longitude and latitude, and assume that the typhoon intensity suffered within each grid is the same.

[0205] In the embodiments of the present application, the typhoon prediction wind speed is combined with the line vulnerability curve to obtain the failure rate of the line in the th grid area, so as to obtain the failure probability of the entire line.

[0206] Among them, the line failure probability model is:

[0207]

[0208] In the formula, is the typhoon wind speed in the th grid area; is the maximum wind speed that the distribution network line can resist; is the empirical coefficient; is the failure probability of the line in the th grid area at time is the total failure probability of the line at time

[0209] The failure probability model of the photovoltaic unit is:

[0210]

[0211] In the formula, is the failure probability of the th PV at time is the failure probability of the PV under normal conditions; is the included angle between the th PV and the wind direction at time is the wind speed at the location of the th PV at time and are respectively the upper and lower limits of the wind speed when the PV fails; is the wind direction change sensitivity coefficient, is the wind direction change rate.

[0212] Step S103: According to the line failure probability and the photovoltaic unit failure probability, generate multiple failure scenarios by Monte Carlo sampling.

[0213] The Monte Carlo method is a stochastic simulation method based on probability statistics, which approximately calculates the solutions of complex problems through a large number of random samplings. In the embodiments of the present application, the Monte Carlo method is used to randomly generate multiple possible fault scenarios according to the joint distribution of the line fault probability and the photovoltaic unit fault probability, and each scenario contains specific line and photovoltaic unit fault conditions. These fault scenarios are used to simulate the actual operation of the distribution network under typhoon weather, so as to comprehensively evaluate the operation risk of the distribution network.

[0214] Step S104: Screen out the fault scenario corresponding to the maximum entropy value of the line fault probability and the photovoltaic unit fault probability from multiple fault scenarios, and determine the typical fault scenario of the distribution network.

[0215] Among them, by respectively extracting N possible initial fault moments from multiple fault scenarios, N possible fault scenarios are constructed to establish a set composed of N system information entropy values, and the probability distribution is fitted. The system information entropy value with the largest occurrence probability represents the most likely distribution network fault scenario. Therefore, a certain extreme fault scenario corresponding to this value is selected as the typical fault scenario. Among them, the calculation of the entropy value is as follows:

[0216]

[0217] In the formula, W is the entropy value, is the set of distribution network lines; is the set of distribution network photovoltaics; and are the fault states of the line and the photovoltaic unit respectively (taking 1 when a fault occurs, otherwise 0).

[0218] In some embodiments, according to the typical fault scenario, the distribution network scheduling optimization model is optimized and solved to obtain the distribution network scheduling optimization plan under the typical fault scenario, including:

[0219] Step S501: Randomly generate multiple distribution network scheduling candidate plans that meet the constraint conditions, and initialize the population using the multiple distribution network scheduling candidate plans;

[0220] Step S502: Perform non-dominated sorting on the individuals in the initialized population and calculate the crowding distance;

[0221] Step S503: Determine the fitness value of each individual according to the objective function corresponding to multiple optimization objectives;

[0222] Step S504: According to the fitness value and the crowding distance, select the individuals with a fitness value greater than the preset fitness threshold and a crowding distance greater than the preset crowding threshold to form a new population;

[0223] Among them, the calculation of the crowding distance is carried out by comparing the densities of other individuals around an individual. The smaller the density, the fewer solutions there are around the individual, the larger the crowding distance of the individual, and the more likely it is to be retained during the optimization process. Through non-dominated sorting, individuals with better fitness values are selected, and at the same time, the crowding distance is considered to avoid premature convergence and maintain the diversity of the population. In each iteration, new solutions are generated through operations such as crossover and mutation of the genetic algorithm, continuously approaching the global optimal solution.

[0224] Step S505: Perform crossover and mutation operations on the new population to generate new offspring individuals;

[0225] Step S506: Combine the new offspring individuals with the parent individuals to form a new population, and repeat the non-dominated sorting of the individuals in the initialized population and calculate the crowding distance until the iteration termination condition is met;

[0226] Step S507: Select the optimal individual from the final population as the optimal solution of the distribution network scheduling optimization model to obtain the distribution network scheduling optimization scheme under typical fault scenarios.

[0227] Among them, the distribution network scheduling optimization scheme includes the active power demand of each node at each moment, the active power output by the EV charging station at the node, the active power output by the MESS at the node, the active power output by the PV at the node, the dynamic incentive electricity price, and the set of critical load nodes.

[0228] It can be understood that the embodiment of the present application uses the third-generation non-dominated sorting genetic algorithm (Non-dominated Sorting Genetic Algorithm III, NSGA-III) to optimize and solve the distribution network scheduling optimization model. After obtaining the distribution network scheduling optimization scheme, the improved alternating direction method of multipliers (Alternating Direction Method of Multipliers, ADMM) algorithm is used for distributed solution to coordinate the real-time power distribution of resources such as EVs, energy storage, mobile energy storage, and photovoltaic, to meet the total power balance and the operation constraints of each device; ADMM ensures the decision consistency and convergence of each subsystem through local sub-problem solving and global variable coordination. Specifically, the global power distribution problem is split into the respective optimization problems of EV, MESS, and PV, dual variables are introduced to coordinate the power constraints between subsystems, and while EV, MESS, and PV perform local optimization respectively, they converge to the optimal solution through global coordination.

[0229] Based on the same inventive concept, the embodiment of the present application also provides a distribution network scheduling system for multi-resource collaboration for implementing the above-mentioned distribution network scheduling method for multi-resource collaboration.

[0230] The solution provided by this system for problem-solving is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the distribution network scheduling system for multi-source resource collaboration provided below can refer to the limitations on the distribution network scheduling method for multi-source resource collaboration in the above text, and will not be elaborated here.

[0231] As Figure 3 shown, an embodiment of this application provides a distribution network scheduling system for multi-source resource collaboration, including:

[0232] A fault scenario determination module 100, configured to collect the operation fault data of each component in the distribution network under the typhoon weather scenario, and determine the typical fault scenario of the distribution network according to the operation fault data;

[0233] An optimization objective determination module 200, with the minimum load loss of the distribution network, the minimum total scheduling cost, and the shortest critical load recovery time as multiple optimization objectives;

[0234] A constraint determination module 300, configured to determine the constraint conditions of the multiple optimization objectives according to the balanced operation states of the distribution network and the EV charging stations;

[0235] A model construction module 400, configured to construct a distribution network scheduling optimization model according to the multiple optimization objectives and the constraint conditions;

[0236] A scheduling optimization module 500, configured to perform optimization solution on the distribution network scheduling optimization model according to the typical fault scenario, and obtain the distribution network scheduling optimization plan under the typical fault scenario.

[0237] In some embodiments, the operation fault data includes the line fault probability and the photovoltaic unit fault probability;

[0238] The fault scenario determination module 100 is configured to:

[0239] Predict the typhoon through a pre-trained typhoon time-varying prediction model according to the historical typhoon wind speed, and obtain the predicted typhoon wind speed at the current prediction moment;

[0240] Respectively construct a line fault probability model and a photovoltaic unit fault probability model, and determine the line fault probability and the photovoltaic unit fault probability according to the line fault probability model and the photovoltaic unit fault probability model, in combination with the predicted typhoon wind speed;

[0241] Generate multiple fault scenarios by Monte Carlo sampling according to the line fault probability and the photovoltaic unit fault probability;

[0242] Select the fault scenario corresponding to the maximum entropy value of the line fault probability and the photovoltaic unit fault probability from the multiple fault scenarios, and determine the typical fault scenario of the distribution network.

[0243] In some embodiments, the system further includes: a prediction model construction module, configured to:

[0244] Collect historical typhoon sample time series data, and extract spatio-temporal features from the historical typhoon sample time series data through a convolutional neural network;

[0245] Input the spatio-temporal features into a long short-term memory network to process the time series relationship in the spatio-temporal features, and obtain a typhoon time-varying prediction model.

[0246] In some embodiments, the objective function corresponding to multiple optimization objectives is:

[0247]

[0248] In the formula, is the weighted load loss rate, is the total dispatching cost, is the critical load restoration time; 、 、 are all weight coefficients;

[0249] Among them,

[0250] In the formula, is the active power lost at node n at time t, is the active power demand at node n at time t, is the load importance weight of node n, is the total number of time periods in the dispatching cycle, is the total number of distribution network nodes;

[0251]

[0252] In the formula, is the EV dispatching cost at time; is the energy storage dispatching cost at time; is the PV unit dispatching cost at time; is the charging station incentive cost;

[0253] Among them,

[0254] In the formula, is the unit discharge loss cost of the battery of EV charging station q, is the node set of EV charging stations, is the active power output by EV charging station q at node at time;

[0255]

[0256] Wherein, is the scheduling cost per unit of the MESS unit; is the time node at which the active power output by the MESS is is the scheduling cost per unit distance, is the scheduling distance; is the number of users participating in the scheduling; m is the index of the users participating in the scheduling;

[0257]

[0258] Wherein, is the operation and maintenance cost per unit output of the photovoltaic unit; is the time node at which the active power output by the PV is;

[0259] Among them,

[0260]

[0261] Wherein, is the dynamic incentive electricity price of node ; is the basic incentive electricity price; is the incentive electricity price adjustment coefficient; is the time node at which the active power output by all EV charging stations is;

[0262]

[0263] Wherein, is the unit time cost for the delayed restoration of the critical node; is the set of critical load nodes; is the duration from power outage to power restoration of node n.

[0264] In some embodiments, the constraint conditions include grid operation constraints and EV charging station scheduling constraints; among them, the grid operation constraints include topological constraints, node power balance constraints, node voltage relaxation constraints, line transmission power limit constraints, node voltage limit constraints, node load shedding limit constraints, and PV operation limit constraints;

[0265] The dispatching constraints of the EV charging station include the road network - power grid coupling constraint, the power limit constraint of the EVs participating in the dispatching, the discharge response rate constraint of users under the incentive electricity price, the balance constraint between the discharge response rate of users under the incentive electricity price and the output of the EV charging station, the charge - discharge power limit constraint of the EVs, the power limit constraint of the EVs, and the dispatching limit constraint of the MESS;

[0266] Among them, the road network - power grid coupling constraint is used to constrain the EVs to reach the specified location in time for charging and discharging during the dispatching process;

[0267] The power limit constraint of the EVs participating in the dispatching is used to ensure that the power of the EVs is not lower than the preset threshold at the end of the dispatching;

[0268] The discharge response rate constraint of users under the incentive electricity price is used to determine the discharge response rate of users under different incentive electricity prices according to the sensitivity of users to the incentive electricity price;

[0269] The balance constraint between the discharge response rate of users under the incentive electricity price and the output of the EV charging station is used to ensure that the output of the EV charging station matches the discharge demand of users.

[0270] In some embodiments, the scheduling optimization module 500 is used for:

[0271] Randomly generate multiple distribution network dispatching candidate solutions that meet the constraint conditions, and initialize the population using the multiple distribution network dispatching candidate solutions;

[0272] Perform non - dominated sorting on the individuals in the initialized population, and calculate the crowding distance;

[0273] Determine the fitness value of each individual according to the objective functions corresponding to multiple optimization objectives;

[0274] According to the fitness value and the crowding distance, select the individuals whose fitness value is greater than the preset fitness threshold and the crowding distance is greater than the preset crowding threshold to form a new population;

[0275] Perform crossover and mutation operations on the new population to generate new offspring individuals;

[0276] Merge the new offspring individuals with the parent individuals to form a new population, and repeat performing non - dominated sorting on the individuals in the initialized population and calculating the crowding distance until the iteration termination condition is met;

[0277] Select the optimal individual from the final population as the optimal solution of the distribution network dispatching optimization model to obtain the distribution network dispatching optimization scheme under typical fault scenarios.

[0278] Such as Figure 4As shown in the figure, an embodiment of the present application provides an electronic device. The electronic device 10 includes a memory 20 and a processor 30. A computer program is stored in the memory 20. When the computer program is executed by the processor 30, the processor 30 is caused to execute the steps of the multi-resource collaborative distribution network scheduling method in the above embodiment.

[0279] An embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed, the steps of the multi-resource collaborative distribution network scheduling method in the above embodiment are implemented.

[0280] An embodiment of the present application provides a computer program product. The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer is caused to execute the steps of the multi-resource collaborative distribution network scheduling method described in the above embodiment.

[0281] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the above-described system, electronic device, computer storage medium, and computer program product can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0282] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects and do not necessarily need to be used to describe a specific order or sequence. It should be understood that such data can be interchanged under appropriate circumstances so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "comprising" and "having" and any variations thereof are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units does not necessarily have to be limited to those steps or units clearly listed, but may include other steps or units not clearly listed or inherent to these processes, methods, products, or devices.

[0283] It should be understood that although the steps in the flowcharts involved in the above embodiments are shown in sequence according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear description in this article, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0284] In several embodiments provided by the present invention, it should be understood that the disclosed systems, electronic devices, computer storage media, computer program products, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division, and there may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, indirect couplings or communication connections of devices or units, and can be in electrical, mechanical or other forms.

[0285] The units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they can be located in one place, or can be distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0286] In addition, the functional units in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated units can be implemented in the form of hardware or in the form of software functional units.

[0287] When the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for executing all or part of the steps of the methods described in the various embodiments of the present invention through a computer device (which may be a personal computer, a server, or a network device, etc.). The foregoing storage medium includes: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (English full name: Read-Only Memory, English abbreviation: ROM), random access memories (English full name: Random Access Memory, English abbreviation: RAM), magnetic disks, or optical discs.

[0288] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A distribution network scheduling method with collaborative multi-source resources, characterized in that, Including: Collecting the operation fault data of each component of the distribution network in the typhoon weather scenario, and determining the typical fault scenario of the distribution network according to the operation fault data; Taking the minimization of load loss of the distribution network, the minimization of total dispatching cost, and the shortest critical load recovery time as multiple optimization objectives; Determining the constraint conditions of the multiple optimization objectives according to the balanced operation state of the distribution network and EV charging stations; Constructing a distribution network dispatching optimization model according to the multiple optimization objectives and the constraint conditions; Optimally solving the distribution network dispatching optimization model according to the typical fault scenario to obtain a distribution network dispatching optimization scheme under the typical fault scenario.

2. The multi-resource coordinated distribution network dispatching method according to claim 1, characterized in that: The operation fault data includes line fault probability and photovoltaic unit fault probability; The collecting the operation fault data of each component of the distribution network in the typhoon weather scenario, and determining the typical fault scenario of the distribution network according to the operation fault data includes: Predicting according to the historical typhoon wind speed through a pre-trained typhoon time-varying prediction model to obtain the predicted typhoon wind speed at the current prediction moment; Respectively constructing a line fault probability model and a photovoltaic unit fault probability model, and determining the line fault probability and the photovoltaic unit fault probability according to the line fault probability model and the photovoltaic unit fault probability model, in combination with the predicted typhoon wind speed; Generating a plurality of fault scenarios by Monte Carlo sampling according to the line fault probability and the photovoltaic unit fault probability; Screening out the fault scenario corresponding to the maximum entropy value of the line fault probability and the photovoltaic unit fault probability from the plurality of fault scenarios, and determining the typical fault scenario of the distribution network.

3. The multi-resource coordinated distribution network dispatching method according to claim 1, characterized in that: Also including: Collecting typhoon historical sample time series data, and extracting spatio-temporal features of the typhoon historical sample time series data through a convolutional neural network; Inputting the spatio-temporal features into a long short-term memory network to process the time series relationship in the spatio-temporal features, and obtaining a typhoon time-varying prediction model.

4. The multi-source resource collaborative distribution network scheduling method according to claim 1, wherein, The objective function corresponding to the multiple optimization objectives is: Where, is the weighted load loss rate, is the total scheduling cost, It is the critical load recovery time; 、 、 All are weight coefficients; Among them, Wherein, is the active power loss of node n at time t, is the active power demand of node n at time t, is the load importance weight of node n, is the total number of time periods in the scheduling cycle, is the total number of distribution network nodes; wherein, is the EV scheduling cost at time is the energy storage scheduling cost at time is the PV unit scheduling cost at time is the charging station incentive cost; Among them, In the formula, is the unit discharge loss cost of the battery of EV charging station q, is the set of EV charging stations at node , is the active power output by EV charging station q at node at time In the formula, is the dispatching cost per unit of MESS; is the active power output by MESS at time node ; is the dispatching cost per unit distance, is the dispatching distance; is the number of users participating in dispatching; m is the index of users participating in dispatching; wherein, is the operation and maintenance cost per unit output of the photovoltaic unit; is the time node at which the active power output by the PV is; Among them, Wherein, is the dynamic incentive electricity price of node ; is the basic incentive electricity price; is the incentive electricity price adjustment coefficient; is the active power output by all EV charging stations at node at time ; Wherein, is the unit time cost for the key node delay recovery; is the key load node set; is the duration from power outage to power restoration of node n.

5. The multi - resource collaborative distribution network scheduling method according to any one of claims 1 to 4, characterized in that The constraint conditions include power grid operation constraints and EV charging station dispatching constraints; among them, the power grid operation constraints include topological constraints, node power balance constraints, node voltage relaxation constraints, line transmission power limit constraints, node voltage limit constraints, node load shedding limit constraints, and PV operation limit constraints; The EV charging station dispatching constraints include road network-power grid coupling constraints, EV power limits participating in dispatching, user discharge response rate constraints under incentive electricity prices, balance constraints between user discharge response rates and EV charging station output under incentive electricity prices, EV charge and discharge power limit constraints, EV power limits, and MESS dispatching limit constraints; Among them, the road network-power grid coupling constraint is used to constrain EVs to reach the designated position in time for charging and discharging during the dispatching process; The EV power limit participating in dispatching is used to ensure that the EV power is not lower than a preset threshold at the end of the dispatching; The user discharge response rate constraint under incentive electricity prices is used to constrain the user discharge response rate under different incentive electricity prices according to the sensitivity of the user to incentive electricity prices; The balance constraint between the discharge response rate of the user under the incentive electricity price and the output of the EV charging station is used to ensure that the output of the EV charging station matches the discharge demand of the user.

6. The multi - resource collaborative distribution network scheduling method according to claim 5, wherein According to the typical fault scenario, the distribution network scheduling optimization model is optimized and solved to obtain the distribution network scheduling optimization plan under the typical fault scenario, including: Randomly generate multiple distribution network scheduling candidate plans that meet the constraint conditions, and use the multiple distribution network scheduling candidate plans to initialize the population; Perform non-dominated sorting on the individuals in the initialized population and calculate the crowding distance; Determine the fitness value of each individual according to the objective function corresponding to the multi-optimization objective; According to the fitness value and the crowding distance, select individuals with a fitness value greater than the preset fitness threshold and a crowding distance greater than the preset crowding distance threshold to form a new population; Perform crossover and mutation operations on the new population to generate new offspring individuals; Merge the new offspring individuals with the parent individuals to form a new population, and repeat the steps of performing non-dominated sorting on the individuals in the initialized population and calculating the crowding distance until the iteration termination condition is met; Select the optimal individual from the final population as the optimal solution of the distribution network scheduling optimization model to obtain the distribution network scheduling optimization plan under the typical fault scenario.

7. A distribution network scheduling system with collaborative multi-source resources, characterized in that, Including: A fault scenario determination module, configured to collect the operation fault data of each component of the distribution network in a typhoon weather scenario, and determine the typical fault scenario of the distribution network according to the operation fault data; An optimization objective determination module, with the minimization of the load loss of the distribution network, the minimization of the total scheduling cost, and the shortest critical load recovery time as the multi-optimization objectives; A constraint determination module, configured to determine the constraint conditions of the multi-optimization objective according to the balanced operation state of the distribution network and the EV charging station; A model construction module, configured to construct a distribution network scheduling optimization model according to the multi-optimization objective and the constraint conditions; A scheduling optimization module, configured to optimize and solve the distribution network scheduling optimization model according to the typical fault scenario to obtain the distribution network scheduling optimization plan under the typical fault scenario.

8. An electronic device, characterized in that, The electronic device includes a memory and a processor. When the computer program stored in the memory is executed by the processor, the processor executes the steps of the distribution network scheduling method for multi-source resource collaboration according to any one of claims 1-6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed, it implements the steps of the distribution network scheduling method for multi-source resource collaboration according to any one of claims 1-6.

10. A computer program product, characterized in that, The computer program product includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions. When the program instructions are executed by a computer, the computer executes the steps of the distribution network scheduling method for multi-source resource collaboration according to any one of claims 1-6.

Images