Method and device for restoring load of urban power grid based on user demand

Abstract

The application provides a kind of city power grid load recovery method and device based on user demand, applicable to electric power system technical field, comprising: obtaining the historical power load data set of each target user of post-disaster city, and the maintenance resource, distribution network topology structure and power equipment parameter set of post-disaster city;For each target user, based on historical load data set, the power consumption of the target user is analyzed, and the power demand of the target user is obtained;The power demand of all target users, the distribution network topology structure of post-disaster city and the power equipment parameter set are input into the pre-constructed distribution network post-disaster recovery model, and the scheduling strategy is solved with strong maintenance timeliness as the goal, to obtain the scheduling strategy;The post-disaster recovery model of distribution network includes maintenance resource scheduling unit;The power resources and maintenance resources of post-disaster city are scheduled based on the scheduling strategy.The application can improve the recovery speed of post-disaster city power grid and reduce economic loss.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of power systems, and particularly relates to a city power grid load recovery method and device based on user demand. BACKGROUND

[0002] City power distribution networks are of vital importance after natural disasters, and directly related to the rapid recovery of urban infrastructure, and also directly affect the continuity of residents' life and urban economic activities. In this process, fully considering the power supply guarantee needs of users is the core of the recovery work. By reasonably allocating human, material and financial resources, maintaining communication with users, updating the recovery progress in time, and formulating emergency measures and long-term planning, the disaster resistance and flexibility of the power distribution network can be improved, and the impact of future disasters on the power supply system can be reduced.

[0003] In existing research on post-disaster recovery, historical data is usually screened, or a time series power curve is obtained through a prediction tool, and the user power demand is described through the time series power curve. Then, a mixed integer linear programming model of post-disaster power distribution network topology optimization is constructed through a power distribution network branch flow model, and a classic optimization solver such as CPLEX (Concert Technology for Extending Linear Programming Extraction) or GUROBI (Gurobi Optimizer) is used to solve a series of optimization problems containing a large number of 0, 1 variables and continuous variables.

[0004] However, in the process of identifying the power demand of key users after the disaster, the standard recurrent neural network cannot fully process the time sequence characteristics in the input data, and will face the problems of gradient disappearance and gradient explosion. The classic optimization solver has the difficulties of low solving efficiency and inability to solve when solving such problems. SUMMARY

[0005] The embodiments of the present application provide a city power grid load recovery method and device based on user demand to solve the problems of gradient disappearance, gradient explosion and low solving efficiency in existing post-disaster recovery methods.

[0006] The present application is realized by the following technical solutions:

[0007] In a first aspect, the embodiments of the present application provide a city power grid load recovery method based on user demand, comprising:

[0008] obtaining a historical power load data set of each target user of the post-disaster city, and a maintenance resource, power distribution network topology structure and power equipment parameter set of the post-disaster city;

[0009] For each target user, based on the historical load data set of the target user, the target user is analyzed for power consumption, and the power consumption demand of the target user is obtained;

[0010] The power consumption demand of all target users, the distribution network topology structure of the post-disaster city and the power equipment parameter set are input into the pre-constructed distribution network post-disaster recovery model, and the solution is solved with strong maintenance timeliness as the target to obtain a scheduling strategy; the distribution network post-disaster recovery model includes a power system operation unit, a topology reconstruction unit and a maintenance resource scheduling unit;

[0011] The power resources and maintenance resources of the post-disaster city are scheduled based on the scheduling strategy.

[0012] In a second aspect, an embodiment of the present application provides a city power grid load recovery device based on user demand, comprising:

[0013] The acquisition module is configured to acquire the historical power load data set of each target user of the post-disaster city, and the maintenance resources, the distribution network topology structure and the power equipment parameter set of the post-disaster city;

[0014] The analysis module is configured to, for each target user, based on the historical load data set of the target user, analyze the power consumption of the target user to obtain the power consumption demand of the target user;

[0015] The solution module is configured to input the power consumption demand of all target users, the distribution network topology structure of the post-disaster city and all parameters in the power equipment parameter set into the pre-constructed distribution network post-disaster recovery model, and solve with strong maintenance timeliness as the target to obtain a scheduling strategy; the distribution network post-disaster recovery model includes a power system operation unit, a topology reconstruction unit and a maintenance resource scheduling unit;

[0016] The scheduling module is configured to schedule the power resources and maintenance resources of the post-disaster city based on the scheduling strategy.

[0017] The embodiment of the present application provides a city power grid load recovery method and device based on user demand, the power consumption demand of the corresponding target user is obtained through the historical load data set of each target user, a distribution network post-disaster recovery model including a power system operation unit, a topology reconstruction unit and a maintenance resource scheduling unit is pre-constructed, all parameters in the power consumption demand of all target users, the distribution network topology structure of the post-disaster city and the power equipment parameter set are input into the pre-constructed distribution network post-disaster recovery model, a scheduling strategy is obtained, and the power resources and maintenance resources of the post-disaster city are scheduled according to the scheduling strategy. The power consumption demand of the target user is clear, and in the process of scheduling, the power consumption demand of the target user is considered by determining the target user, and strong maintenance timeliness is taken as the target, which can effectively improve the speed of the post-disaster city power grid recovery and further reduce economic losses. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for restoring urban power grid load based on user needs, provided in an embodiment of the present invention.

[0020] Figure 2 This is a schematic diagram illustrating an application scenario of a user-demand-based urban power grid load restoration method according to an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the target user post-disaster electricity demand prediction model based on LSTM, which is provided in an embodiment of the urban power grid load recovery method based on user demand according to an embodiment of the present invention.

[0022] Figure 4 This is a flowchart of the Benders algorithm for solving the main problem and subproblems in a user-demand-based urban power grid load restoration method according to an embodiment of the present invention.

[0023] Figure 5 This is a schematic diagram of the urban power grid load restoration device based on user needs provided in an embodiment of the present invention. Detailed Implementation

[0024] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of the invention. However, those skilled in the art will understand that the invention can be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods are omitted so as not to obscure the description of the invention with unnecessary detail.

[0025] Figure 1 This is a flowchart illustrating a user-demand-based urban power grid load restoration method according to an embodiment of the present invention. (Refer to...) Figure 1 The following is a detailed description of this urban power grid load restoration method based on user demand:

[0026] S110: Obtain the historical power load dataset for each target user in the disaster-stricken city, as well as the maintenance resources, distribution network topology, and power equipment parameter set of the disaster-stricken city.

[0027] Optionally, the target user refers to a post-disaster key protection user, which is confirmed in advance.

[0028] It should be noted that after the occurrence of extreme time, phenomena such as power line breakage and tower collapse may occur, which may cause the interruption of power transmission channels and affect power supply. Referring to Figure 2 , it is necessary to establish emergency management centers, overall command and decision-making centers, repair centers, weather centers, traffic centers, data sharing centers and other functional department centers in emergency situations.

[0029] Among them, the repair center is used to collect fault conditions and repair resource conditions; the weather center is used to report post-disaster weather warning information, and the weather conditions will affect the progress of the repair process; the traffic center is used to master the real-time traffic conditions of the traffic network, and to provide guidance for the traffic path of the repair team.

[0030] These functional department centers will pass information to the data sharing center to provide information data support for the decision-making center to develop post-disaster recovery strategies, wherein the data sharing center needs to ensure the security of shared data and maintain the privacy of key core data of each department.

[0031] The decision-making center is used to decompose the large-scale mixed integer linear programming optimization problem after obtaining the information, iteratively solve the main and sub-problems after decomposition, and feed back the related optimization results to the data sharing center. Each department center cooperates to complete the task of post-disaster recovery of urban power grid according to the optimization results of the data sharing center, so as to achieve the purpose of quickly restoring power supply and reducing economic losses.

[0032] In S120, for each target user, based on the historical load data set of the target user, the electricity consumption of the target user is analyzed to obtain the electricity demand of the target user.

[0033] In this embodiment, the electricity consumption of the target user is analyzed by an LSTM (Long Short Term Memory) neural network model. The electricity consumption of the target user can also be analyzed by other means, such as an ARIMA (Auto Regressive Integrated Moving Average) model, which is not specifically limited in this application.

[0034] In one possible implementation, for each target user, based on the target user's historical load dataset, an electricity consumption analysis is performed to obtain the target user's electricity demand. This includes: for each target user, performing the following steps: obtaining the power influencing factors of the city after the disaster; constructing the target user's feature vector based on the power influencing factors and the target user's historical load dataset; and inputting the target user's feature vector into a pre-built LSTM network model to obtain the target user's electricity demand.

[0035] Among them, the factors affecting electricity include meteorological factors and disaster factors. Meteorological factors mainly include minimum temperature, air quality, sunshine hours, working days, and weather type, while disaster factors mainly include the degree of disaster.

[0036] Optionally, for each target user, the input to the LSTM neural network model is the feature vector of that target user, as shown below:

[0037]

[0038] in, For the first i Target users t The feature vector at time step, The lowest temperature, For air quality, This refers to the number of hours of sunshine. For working days Weather type As for the degree of disaster, For the first i Historical load data for each target user.

[0039] The output of the LSTM neural network model is the target user's... t The electricity demand at all times.

[0040] See Figure 3 The network structure of the LSTM neural network model mainly includes an input layer, a hidden layer, and an output layer. The input of the input layer is the feature vector of the target user to be predicted. The hidden layer contains several nodes of the LSTM. The core of the LSTM design is the threshold mechanism, including the input gate, the forget gate, the memory unit, and the output gate.

[0041] The input gate is mainly used for filtering new information. It determines the information input into the memory cell and updates the current state of the memory cell, deciding which information from the user's electricity demand and habits in the previous moment should be retained. Its mathematical model is shown below:

[0042]

[0043]

[0044] wherein, is the input of the input gate at time t; σ is the Sigmoid function; and is the network weight of the input gate, is the bias of the input gate, is t user electricity information retained at time -1; is the candidate memory cell state at time t; and is the network weight of the memory cell, is the bias of the memory cell.

[0045] The forget gate is used to control the influence of the memory of the previous time step on the current time step, and it decides which user electricity information should be discarded, and its mathematical model is:

[0046]

[0047] wherein, is t the input of the forget gate at time t; and is the network weight of the forget gate, is the bias of the forget gate, and the value of the LSTM forget gate can be selected between [0, 1] to achieve the use of LSTM to improve the gradient vanishing and gradient explosion in the process of processing user electricity demand information.

[0048] The memory cell state is similar to a conveyor belt, which runs through the entire LSTM neural model, and its role is to store the network information of the current time LSTM and pass it down. The calculation formula of the memory cell state is as follows:

[0049]

[0050] wherein, is the memory cell state at time t-1; is the updated memory cell state at time t after passing through the input gate and the forget gate.

[0051] The output gate determines the final output of the LSTM neural network model and the retained user power consumption information. The output gate uses a sigmoid function to weight and add the input information at the current time and the output information at the previous time, to obtain an initial output. The initial output is scaled by a tanh activation function to obtain memory information learned before, and then multiplied by the initial output to obtain the output of the LSTM neural network model, i.e., the power consumption demand of the ith target user at the t time. The mathematical model of the output gate is as follows:

[0052]

[0053]

[0054] wherein, is the state information retained at the t time; and is the weight of the output gate network, is the bias of the output gate.

[0055] The hidden layer adjusts the related parameters in the mathematical model by using the gradient calculated by the loss function and back propagation, i.e., uses the Adam algorithm to update the model parameters, and iteratively optimizes until the loss function converges. The number of hidden layers in the LSTM model, the number of units in each layer, the maximum number of iterations, the initial learning rate, the number of iterations when the learning rate decreases, and the learning rate reduction factor are optimized by using the particle swarm algorithm. After the model training is completed, the result is processed by the output layer to obtain the post-disaster user power consumption demand.

[0056] By introducing the LSTM to process the historical power consumption data of the target user, and by using the gating mechanism to selectively retain the features of the data in the transmission process, the information can be long-term maintained and updated, so that the neural network can more accurately analyze the differentiated power consumption characteristics of the target user, and further determine the post-disaster power consumption demand.

[0057] S130, input the power consumption demand of all target users, the power distribution network topology structure of the post-disaster city, and the power equipment parameter set into the pre-constructed power distribution network post-disaster recovery model, and solve with the goal of strong repair timeliness to obtain a scheduling strategy. The power distribution network post-disaster recovery model includes a power system operation unit, a topology reconstruction unit, and a repair resource scheduling unit.

[0058] In one possible implementation, the repair resources of the post-disaster city include repair team resources of the post-disaster city, path resources of the post-disaster city, and traffic resources of the post-disaster city; and the power equipment parameter set includes device parameters of normally operating power equipment and device parameters of faulty power equipment.

[0059] Optionally, electrical equipment refers to equipment used in distribution networks or microgrids for power supply, transmission or transformation, excluding electrical equipment used by the target users of the electrical equipment.

[0060] Optionally, the power system operation unit in the distribution network post-disaster recovery model mainly consists of distribution network operation constraints and microgrid operation constraints; the topology reconfiguration unit is mainly obtained by representing the radial distribution network topology model with a virtual power balance constraint model; the maintenance resource scheduling unit can ensure that each damaged component can be repaired by a maintenance team, and the maintenance team arriving at the node can leave on its own after completing the maintenance work. All maintenance teams can depart from the maintenance site, and the time for the maintenance team to arrive at the damaged component and the repair status of the damaged component in each time period are constrained.

[0061] Optionally, the distribution network operation constraints in the power system operation unit may include distribution network node power balance constraints, power factor constraints that should be met when active and reactive loads are removed, branch power flow constraints that should be met when power on the line and voltage at each node, upper and lower limit constraints on voltage at each node, and upper and lower limit constraints on power output of microgrids and the upper-level grid.

[0062] The power balance constraints at distribution network nodes are:

[0063]

[0064] The power factor constraints that must be satisfied when disconnecting active and reactive loads are:

[0065]

[0066] The branch power flow constraints that the power and voltage at each node on the line must satisfy are as follows:

[0067]

[0068]

[0069] The upper and lower limits of the voltage at each node are:

[0070]

[0071] The upper and lower limits of the power output of the upstream power grid are constrained as follows:

[0072]

[0073] in, For the upper-level power grid t Node of time i Active power output at the location, For the upper-level power grid t Time Nodei reactive power output at the location For microgrids in t Node of time i Active power output at the location, For microgrids in t Time Node i reactive power output at the location for t Time Node i Active load removed at the site for t Time Node i The reactive load removed at the site, For nodes i The set of connected downlinks, For nodes i The set of connected uplink lines, For target users t Time Node i The electricity demand at the location For target users t Time Node i reactive load at the location, In order to be in t Time Node i The power factor of the load, branch road l exist t Active power at any given time branch road l exist t Reactive power at any given moment branch road l exist t Maximum apparent power at any given time For nodes i exist t Voltage value at time, For node j in t Voltage value at time, branch road l The resistance, branch road l Reactance The system reference voltage value, M is a large number related to the reference value of the system node voltage, and the line... l exist t When there are no faults ,line l exist t When a fault is always present , For upper limit constraints on power output of the upper-level power grid and microgrid, As a lower limit constraint for the power output of the upper-level power grid and microgrid, The active or reactive power output of the upper-level power grid and the microgrid at the time point t. t The active or reactive power output of the upper-level power grid and the microgrid at the time point t. i The active or reactive power output of the upper-level power grid and the microgrid at the time point t.

[0074] The microgrid operation constraints in the power system operation unit include upper and lower limit constraints of distributed generator output power, upper and lower limit constraints of energy storage system charging and discharging power, total quantity balance constraints after load transfer in the microgrid, and line power balance constraints in the microgrid.

[0075] The upper and lower limit constraints of the distributed generator output power are as follows:

[0076]

[0077] The upper and lower limit constraints of the energy storage system charging and discharging power are as follows:

[0078]

[0079]

[0080]

[0081] The total quantity balance constraints after load transfer in the microgrid are as follows:

[0082]

[0083]

[0084]

[0085]

[0086]

[0087] The line power balance constraints in the microgrid are as follows:

[0088]

[0089] wherein, is the output power of the distributed generator at the time point t, i is the output power of the distributed generator at the time point t, t is the upper limit of the distributed generator output power, is the lower limit of the distributed generator output power, is the charging power of the energy storage system at the time point t, is the charging power of the energy storage system at the time point t, j is the discharging power of the energy storage system at the time point t, t is the discharging power of the energy storage system at the time point t, is the discharging power of the energy storage system at the time point t, j is the discharging power of the energy storage system at the time point t, t is the discharging power of the energy storage system at the time point t, Lower limit of charging power of energy storage system, Upper limit of charging power of energy storage system, Lower limit of discharging power of energy storage system, Upper limit of discharging power of energy storage system, Energy storage system j Charging power coefficient at time t , Energy storage system j Discharging power coefficient at time t , Energy storage system j Lower limit of state of charge of energy storage system, Energy storage system j Upper limit of state of charge of energy storage system, Discharging efficiency coefficient, Charging efficiency coefficient, Energy storage system j Rated energy value of energy storage system, Time step, And N PCC Interconnection line l Input power and set at time t , N cDG Set of distributed generators, N ESS Set of energy storage systems, N DR Set of controllable loads participating in demand response, Controllable load m Power at time t , Load shifting amount of controllable load in a period, Lower limit of load shifting amount of controllable load at time t , Upper limit of load shifting amount of controllable load at time t , And Controllable load m Auxiliary variable of controllable load t Joining at time , m Scheduling cost of controllable load t At time .

[0090] The topology reconstruction unit in the post-disaster recovery model of the power distribution network is:

[0091]

[0092]

[0093]

[0094] in, For power transmission lines l At any moment t Virtual power flow, For the number of nodes in the power system, To ensure the 0 / 1 variables of the radial structure of the distribution network, Indicates the first i Are the potential root nodes in...? t The set of potential root nodes that are selected as the root node at any given time. R For the damaged nodes at both ends of the distribution network line, For nodes i The set of connected downlink / uplink lines, ε This refers to a collection of distribution network lines.

[0095] In one possible implementation, the maintenance resource scheduling unit includes:

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104]

[0105] in, For repair stations and the collection of damaged components, In c For the site The maintenance teams under its jurisdiction assembled. A 0 / 1 variable used to represent the repair team. c Was a route selected? For the repair team to remove damaged components d Drive to the damaged component s , The repair team failed to remove the damaged components. dtravel to the damaged element s , = 1 if the maintenance team travels through the damaged element d = 0 if the maintenance team does not travel through the damaged element d , d = 1 if the maintenance team travels through the damaged element = 0 if the maintenance team does not travel through the damaged element s , = 1 if the maintenance team travels through the damaged element d = 0 if the maintenance team does not travel through the damaged element = 1 if the maintenance team travels through the damaged element c = 0 if the maintenance team does not travel through the damaged element d = 1 if the maintenance team travels through the damaged element = 0 if the maintenance team does not travel through the damaged element d = 1 if the maintenance team travels through the damaged element s = 0 if the maintenance team does not travel through the damaged element = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time t , t = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time , t = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time , ε = 1 if the damaged element is being repaired at time = 0 if the damaged element is not being repaired at time t , = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time k , k = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time , k = 1 if the damaged element is being repaired at time d = 0 if the damaged element is not being repaired at time , = 1 if the damaged element is being repaired at time

[0106] In one possible implementation, the electricity demand of all target users, the distribution network topology of the city after the disaster, and the parameter set of power equipment are input into a pre-constructed distribution network post-disaster recovery model. The model is then solved with the goal of achieving high maintenance timeliness to obtain a scheduling strategy. This includes: inputting the electricity demand of all target users, the distribution network topology of the city after the disaster, and the parameter set of power equipment into the distribution network post-disaster recovery model to obtain a mixed-integer linear programming model; constructing the objective function and related constraints of the mixed-integer linear programming model with the goals of minimizing post-disaster load shedding losses, reducing microgrid operating costs, and achieving high maintenance timeliness; transforming the objective function and related constraints into a compact form; and decomposing the mixed-integer linear programming model into a master model and sub-models based on the Benders algorithm; and solving the master model and sub-models to obtain the scheduling strategy.

[0107] In one possible implementation, the objective function is:

[0108]

[0109]

[0110] in, i For the distribution network i 1 node t For a moment, j For the first j One energy storage system, l For connecting the distribution network and the microgrid, This is the unit coefficient for load shedding loss. For the operating cost of distributed power sources, For the operating costs of energy storage systems, For demand response costs, To cover the cost of purchasing and selling electricity, This refers to load importance, which is related to electricity demand. This represents the weighting coefficient for the entire maintenance process. For the set of nodes in the power system, Gathering at the repair station A collection of damaged components. It is a collection of distributed generators. A collection of energy storage devices, For the set of loads participating in demand response, It is a collection of connecting lines between the distribution network and the microgrid. This represents the operating cost coefficient for distributed power sources. and The coefficients of the first-order and constant terms in the expression for converting the output power of a distributed power source are... This is the operating cost coefficient for the energy storage system. a demand response cost, a t a demand response power reduced due to load shedding at the moment, a cost coefficient of power purchase of the distribution network from the upper-level power grid, a cost coefficient of power sale of the distribution network to the user, a power purchase amount of the distribution network from the upper-level power grid, a power sale amount of the distribution network to the user.

[0111] The target function is converted into a compact form as follows:

[0112]

[0113] The relevant constraints are converted into a compact form as follows:

[0114]

[0115]

[0116] wherein, and are discrete and continuous decision variables, that is, dispatching instructions of various post-disaster recovery resources of the urban distribution network, the micro-grid and the repair team, and are vector forms of dual variables associated with the continuous and discrete decision variable equality and inequality constraints, and are coefficient vectors of the discrete variable vector and the continuous variable vector, A , D are coefficient matrices of the equality or inequality constraints containing 0 / 1 binary variables in the compact form, and are coefficient vectors corresponding to the equality or inequality constraints of the discrete variables, E , G and H are coefficient matrices of the equality or inequality constraints containing continuous variables, is a coefficient vector corresponding to the equality or inequality constraints of the continuous variables, is a coefficient vector corresponding to the equality or inequality constraints satisfied by both the continuous variables and the discrete variables.

[0117] In a possible implementation, the main model and the sub-model are solved to obtain the dispatching strategy, including: converting the sub-model into a dual expression and relevant constraints of the dual expression; based on the main model, the dual expression and the relevant constraints of the dual expression, iteratively solving to obtain an optimal dispatching strategy.

[0118] Optionally, in constructing the objective function, each target, such as small post-disaster load shedding loss, small micro-grid operation cost, strong timeliness in maintenance, etc., corresponds to a sub-model, and at the same time, the sub-model corresponding to each target is similar.

[0119] Optionally, the objective function of the sub-model contains the dual variables associated with the related constraints in the post-disaster recovery model of the urban distribution network and the discrete variables of the decision variables of the post-disaster recovery model of the urban distribution network.

[0120] In a possible implementation, the sub-model is:

[0121]

[0122]

[0123]

[0124] wherein, is the discrete variable decision result of the post-disaster recovery model of the urban distribution network passed from the main problem, t is the time point, and is the vector form of the dual variables associated with the equality and inequality constraints of the continuous and discrete decision variables, q is the optimal objective function value, E , G , H is the coefficient matrix of the equality or inequality constraint containing the continuous variable, is the coefficient vector corresponding to the equality or inequality constraint of the continuous variable.

[0125] wherein, the objective function of the sub-model at any time point is:

[0126]

[0127] It should be noted that the corresponding constraint of the sub-model is the constraint that the dual variables associated with the corresponding constraint in the post-disaster recovery model of the distribution network should satisfy, and the mathematical relationship of the dual expression needs to be satisfied. The related constraint of the sub-model is as follows:

[0128]

[0129] Optionally, the objective function of the main model is composed of the discrete variables of the decision variables of the post-disaster recovery model of the distribution network and the optimal objective function value of the sub-problem. q

[0130] In a possible implementation, the main model is:

[0131] ​

[0132]

[0133]

[0134] wherein, t is the time, k is the iteration number, and is the sub-model scheduling decision variable of the post-disaster recovery resources at time t and the k th iteration, q is the optimal objective function value, is the scheduling instruction of the post-disaster recovery resources of the urban distribution network, the micro-grid, and the repair team, is the coefficient vector of the discrete variable vector, A , D is the coefficient matrix of the equation or inequality constraint containing the 0 / 1 binary variable in the compact form, and is the coefficient vector corresponding to the equation or inequality constraint of the discrete variable, G is the coefficient matrix of the equation or inequality constraint containing the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint of the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint satisfied by the continuous variable and the discrete variable.

[0135] wherein the objective function of the master model is:

[0136]

[0137] wherein the first constraint of the master model is:

[0138]

[0139] the second constraint of the master model is:

[0140]

[0141] the third constraint of the master model is:

[0142]

[0143] Referring to Figure 4 , the solving process is as follows:

[0144] First, the objective function of the relaxed master model is solved, at this time, the last two constraints of the master model do not need to be added, and a set of solutions ( , ) is obtained.

[0145] Secondly, the set of solutions is substituted into the sub-model, and the values of and are calculated, and and are determined as unbounded solutions or bounded solutions. If and are unbounded solutions, that is, the dual variables associated with the corresponding constraints in the post-disaster restoration model of the power distribution network cannot be obtained, a third constraint of the master model, that is, the lower bound of the range of the post-disaster restoration model of the power distribution network, needs to be added to the master model, and the objective function of the relaxed master model needs to be recalculated;

[0146] If and are bounded solutions, that is, , the extreme point of the dual sub-problem, that is, the candidate optimal solution, can be obtained.

[0147] Finally, the values of and are compared. If they are equal, the loop is terminated, and the solution at this time is output. If they are not equal, the optimal cut constraint, that is, the second constraint of the master model, that is, the upper bound of the range of the post-disaster restoration model of the power distribution network, needs to be added, and the objective function of the master model needs to be recalculated. The iteration loop is performed until the upper bound and the lower bound of the post-disaster restoration model of the power distribution network are constantly approaching, and the optimal solution is obtained.

[0148] It should be noted that the optimal solution refers to the optimal dispatching strategy of various resources such as the urban power distribution network, the micro-grid, and the repair team.

[0149] Since the number of bounded solutions and unbounded solutions generated by the sub-model in the process of solving the post-disaster restoration model of the power distribution network is limited, and new feasibility constraints and optimality constraints are generated in each iteration, the constraint conditions in the master model of the post-disaster restoration model of the power distribution network are updated, therefore, after a finite number of iterations, the obtained solution will converge to the optimal solution, that is, the optimal dispatching strategy of various resources such as the urban power distribution network, the micro-grid, and the repair team will be solved.

[0150] By constructing a mixed integer linear programming model for the topology optimization of the power distribution network, and constructing the objective function and related constraints of the mixed integer linear programming model for the topology optimization of the power distribution network with the objectives of small post-disaster load shedding, low micro-grid operation cost, and strong timeliness of repair, the post-disaster power distribution network can be quickly reconstructed. The scheduling scheme is quickly solved by the efficient Benders decomposition algorithm. The optimization problem containing a large number of decision variables is decomposed into a master model and a sub-model, which is helpful to quickly and accurately find the optimal solution of the model, that is, the optimal dispatching strategy.

[0151] S140 is used to schedule power and maintenance resources in disaster-stricken cities based on scheduling strategies.

[0152] Optionally, after obtaining the optimal scheduling strategy, through Figure 2 The decision-making center, as shown, feeds back the optimal dispatch strategy to the data sharing center. Each department obtains its corresponding departmental dispatch strategy through the data sharing center and carries out post-disaster recovery work on the urban power grid according to the dispatch strategy. The urban distribution network dispatches its own resources to restore the distribution network load according to the instructions corresponding to the optimal dispatch strategy; the microgrid, in turn, dispatches its resources to assist the distribution network in load restoration according to the instructions corresponding to the optimal dispatch strategy; maintenance teams repair lines according to the path-related strategies in the optimal dispatch strategy and promptly upload repair information to the data sharing center, providing a reference for the overall power supply recovery of the urban distribution network.

[0153] By obtaining the electricity demand of each target user from their historical load dataset, a post-disaster recovery model for the distribution network is pre-constructed, including power system operation units, topology reconfiguration units, and maintenance resource scheduling units. The electricity demand of all target users, the distribution network topology of the post-disaster city, and all parameters from the power equipment parameter set are input into the pre-constructed model to obtain a scheduling strategy. Based on this strategy, power and maintenance resources in the post-disaster city are then scheduled. This approach clearly identifies the electricity demand of target users. By determining and considering their electricity needs, and with objectives such as high maintenance timeliness, minimal post-disaster load shedding losses, and low microgrid operating costs, the efficient Benders decomposition algorithm is used to quickly solve the scheduling scheme. This effectively accelerates the recovery speed of the post-disaster urban distribution network and reduces economic losses and maintenance costs.

[0154] It should be understood that the sequence number of each step in the above embodiments does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0155] Corresponding to the user-demand-based urban power grid load restoration method in the above embodiments, Figure 5 The diagram shows a structural schematic of an urban power grid load restoration device based on user needs provided by an embodiment of the present invention. For ease of explanation, only the parts related to the embodiment of the present invention are shown.

[0156] See Figure 5 The urban power grid load restoration device 5 based on user needs in this embodiment of the invention may include:

[0157] The acquisition module 51 is configured to acquire a historical power load data set of each target user in the post-disaster city, and a set of maintenance resources, a power distribution network topology and power equipment parameters of the post-disaster city.

[0158] The analysis module 52 is configured to perform power consumption analysis on each target user based on the historical load data set of the target user, to obtain power consumption demand of the target user.

[0159] The solving module 53 is configured to input the power consumption demand of all target users, the power distribution network topology of the post-disaster city and the set of power equipment parameters into a pre-constructed post-disaster recovery model of the power distribution network, and solve the post-disaster recovery model with high maintenance timeliness as an objective, to obtain a scheduling strategy. The post-disaster recovery model of the power distribution network includes a power system operation unit, a topology reconfiguration unit and a maintenance resource scheduling unit.

[0160] The scheduling module 54 is configured to schedule the power resources and the maintenance resources of the post-disaster city based on the scheduling strategy.

[0161] In a possible implementation, the analysis module 52 is specifically configured to: for each target user, perform the following steps: acquire power influence factors of the post-disaster city, construct a feature vector of the target user based on the power influence factors and the historical load data set of the target user; and input the feature vector of the target user into a pre-constructed LSTM network model, to obtain the power consumption demand of the target user.

[0162] In a possible implementation, the set of maintenance resources of the post-disaster city includes maintenance team resources of the post-disaster city, path resources of the post-disaster city and traffic resources of the post-disaster city; and the set of power equipment parameters includes device parameters of normally operating power equipment and device parameters of faulty power equipment.

[0163] In a possible implementation, the maintenance resource scheduling unit includes:

[0164]

[0165]

[0166]

[0167]

[0168]

[0169]

[0170]

[0171]

[0172]

[0173]

[0174] in, For repair stations and the collection of damaged components, In c For the site The maintenance teams under its jurisdiction assembled. A 0 / 1 variable used to represent the repair team. c Was a route selected? For the repair team to remove damaged components d Drive to the damaged component s , The repair team failed to remove the damaged components. d Drive to the damaged component s , Does the repair team pass through the damaged component? d If the repair team passes by the damaged component d If the repair team does not reach the damaged component, the value will be 1. d Then the value is 0. For the repair team to reach the damaged components s The time at the location For the repair team to reach the damaged components d The time at the location For the repair team c Repair damaged components d The time required For the repair team to remove damaged components d Location to damaged component s The required travel time to the current location. A 0 / 1 variable used to represent a damaged component. d Is it in t It is constantly being repaired, if t At any moment, damaged components d It is being repaired. ,like t At any moment, damaged components d If the repair work has not yet started or has already been completed, then , ε It is a decimal, related to the allowable maintenance time error precision of the maintenance team scheduling model. To t The number of damaged components that have been repaired so far. A 0 / 1 variable used to represent a damaged component. d Is it ink It is constantly being repaired, if k At any moment, damaged components d It is being repaired. =1, if k At any moment, damaged components d If the repair work has not yet started or has already been completed, then =0.

[0175] In one possible implementation, the solution module 52 is specifically used to: input the electricity demand of all target users, the distribution network topology of the city after the disaster, and the set of power equipment parameters into the distribution network post-disaster recovery model to obtain a mixed-integer linear programming model; and construct the objective function and related constraints of the mixed-integer linear programming model with the objectives of minimizing post-disaster load shedding losses, reducing microgrid operating costs, and improving maintenance timeliness; transform the objective function and related constraints into a compact form, and decompose the mixed-integer linear programming model into a main model and sub-models based on the Benders algorithm; and solve the main model and sub-models to obtain the scheduling strategy.

[0176] In one possible implementation, the objective function is:

[0177]

[0178]

[0179] in, i For the distribution network i 1 node t For a moment, j For the first j One energy storage system, l For connecting the distribution network and the microgrid, This is the unit coefficient for load shedding loss. For the operating cost of distributed power sources, For the operating costs of energy storage systems, For demand response costs, To cover the cost of purchasing and selling electricity, This refers to load importance, which is related to electricity demand. This represents the weighting coefficient for the entire maintenance process. For the set of nodes in the power system, Gathering at the repair station A collection of damaged components. It is a collection of distributed generators. A collection of energy storage devices, For the set of loads participating in demand response, It is a collection of connecting lines between the distribution network and the microgrid. a distributed power supply operation cost coefficient, and a first-order term and a constant term coefficient of a distributed power supply output power conversion expression, an energy storage system operation cost coefficient, a demand response cost, a t a demand response power reduced due to load shedding at the moment, a power grid purchase cost coefficient of the distribution network, a power grid sale cost coefficient of the distribution network, a power grid purchase amount of the distribution network, a power grid sale amount of the distribution network.

[0180] In a possible implementation, the solving module 52 is further configured to: convert the sub-model into a dual expression and related constraints of the dual expression; and perform iterative solving based on the master model, the dual expression and the related constraints of the dual expression to obtain the optimal scheduling strategy.

[0181] In a possible implementation, the sub-model is:

[0182]

[0183]

[0184]

[0185] wherein, a discrete variable decision result of a city distribution network post-disaster recovery model passed from the master problem, t a moment, and a vector form of a dual variable associated with the continuous and discrete decision variable equality and inequality constraints, q an optimal objective function value, E 、 G 、 H a coefficient matrix of an equality or inequality constraint containing a continuous variable, a coefficient vector corresponding to the equality or inequality constraint of the continuous variable.

[0186] In a possible implementation, the master model is:

[0187]

[0188]

[0189]

[0190] wherein,t is the time, k is the iteration number, and is the vector form of the dual variables of the post-disaster recovery resource scheduling decision variables at time t and the k iteration, q is the optimal objective function value, is the scheduling instruction of various post-disaster recovery resources of the urban distribution network, micro-grid, and repair team, is the coefficient vector of the discrete variable vector, A , D is the coefficient matrix of the equation or inequality constraint containing the 0 / 1 binary variable in the compact form, and is the coefficient vector corresponding to the equation or inequality constraint of the discrete variable, G is the coefficient matrix of the equation or inequality constraint containing the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint of the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint satisfied by the continuous variable and the discrete variable.

[0191] In the above embodiments, the description of each embodiment has its own focus, and the parts not described or recorded in a certain embodiment can be referred to the related description of other embodiments.

[0192] Those of ordinary skill in the art can realize that the templates, units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software manner depends on the specific application and design constraints of the technical solution. The skilled person can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0193] The modules / units, if realized in the form of software function units and sold or used as independent products, can be stored in a computer readable storage medium. Based on such understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a computer readable storage medium. When the computer program is executed by a processor, the steps of each of the above-mentioned various city power grid load recovery methods based on user demand can be implemented. The computer program includes computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The computer readable medium can include any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal and software distribution medium, etc.

[0194] The above-mentioned embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for load restoration of urban power grid based on user demand, characterized in that, The method comprises the following steps: obtaining a historical power load data set of each target user in a post-disaster city, and a set of maintenance resources, power distribution network topology and power equipment parameters of the post-disaster city; for each target user, performing power consumption analysis on the target user based on the historical load data set of the target user to obtain power consumption demand of the target user; inputting the power consumption demand of all target users, the power distribution network topology of the post-disaster city and the set of power equipment parameters into a pre-constructed post-disaster recovery model of the power distribution network, and solving the model with the goal of strong maintenance timeliness to obtain a scheduling strategy; the post-disaster recovery model of the power distribution network comprises a power system operation unit, a topology reconstruction unit and a maintenance resource scheduling unit; scheduling the power resources and the maintenance resources of the post-disaster city based on the scheduling strategy; the step of inputting the power consumption demand of all target users, the power distribution network topology of the post-disaster city and the set of power equipment parameters into a pre-constructed post-disaster recovery model of the power distribution network, and solving the model with the goal of strong maintenance timeliness to obtain a scheduling strategy, comprises: inputting the power consumption demand of all target users, the power distribution network topology of the post-disaster city and the set of power equipment parameters into the post-disaster recovery model of the power distribution network to obtain a mixed integer linear programming model, and constructing an objective function and related constraints of the mixed integer linear programming model with the goals of small post-disaster load shedding loss, small micro-grid operation cost and strong maintenance timeliness; transforming the objective function and the related constraints into a compact form, decomposing the mixed integer linear programming model into a master model and a sub-model based on Benders algorithm; solving the master model and the sub-model to obtain a scheduling strategy.

2. The method for load restoration of urban power grid based on user demand according to claim 1, characterized in that, the objective function is: in, i For the distribution network i 1 node t For a moment, j For the first j One energy storage system, l For connecting the distribution network and the microgrid, This is the unit coefficient for load shedding loss. For the operating cost of distributed power sources, For the operating costs of energy storage systems, For demand response costs, To cover the cost of purchasing and selling electricity, The load importance is related to electricity demand. This represents the weighting coefficient for the entire maintenance process. For the set of nodes in the power system, Gathering at the repair station Collection of damaged components It is a collection of distributed generators. A collection of energy storage devices, For the set of loads participating in demand response, It is a collection of connecting lines between the distribution network and the microgrid. This represents the operating cost coefficient for distributed power sources. and The coefficients of the first-order and constant terms in the expression for converting the output power of a distributed power source are... This is the operating cost coefficient for the energy storage system. For demand response costs, for t The demand response power is reduced due to load shedding at all times. The cost coefficient for the distribution network to purchase electricity from the superior power grid. This refers to the cost coefficient for electricity sold from the distribution network to users. The electricity purchased by the distribution network from the superior power grid. Electricity sold from the distribution network to users.

3. The method for load restoration of urban power grid based on user demand according to claim 1, characterized in that, the step of solving the master model and the sub-model to obtain a scheduling strategy, comprises: transforming the sub-model into a dual expression and related constraints of the dual expression; iteratively solving the master model, the dual expression and the related constraints of the dual expression to obtain an optimal scheduling strategy.

4. The method for load restoration of urban power grid based on user demand according to claim 1, characterized in that, the maintenance resource scheduling unit comprises: wherein, is a set of repair sites and a set of damaged elements, c is a set of sites under the jurisdiction of a set of repair teams, is a 0 / 1 variable indicating whether a repair team c has made a route selection, is the time for a repair team to travel from the location of a damaged element d to the location of the damaged element s , is the time for a repair team to travel from the location of a damaged element d to the location of the damaged element s , is a 0 / 1 variable indicating whether a repair team passed by a damaged element d , if the repair team passed by the damaged element d , then it takes the value 1, if the repair team did not pass by the damaged element d , then it takes the value 0, is the time for a repair team to arrive at the location of a damaged element s , is the time for a repair team to arrive at the location of a damaged element d , is the time for a repair team to repair a damaged element c , d is the travel time for a repair team to travel from the location of a damaged element d to the location of the damaged element s , is a 0 / 1 variable indicating whether a damaged element d was repaired at t , if at t the damaged element d was being repaired, then , if at t the repair of the damaged element d had not started or had been completed, then , ε is a small number related to the precision of the repair time error allowed by the repair team scheduling model, is the number of damaged elements that had been repaired by t , is a 0 / 1 variable indicating whether a damaged element d was repaired at k , if at k the damaged element d was being repaired, then = 1, if at k the repair of the damaged element​​ d the repair work has not yet started or has been completed, = 0.

5. The method for load restoration of urban power grid based on user demand according to claim 1, characterized in that, the step of performing power consumption analysis on each target user based on the historical load data set of the target user to obtain power consumption demand of the target user, comprises: for each target user, performing the following steps: obtaining power influence factors of the post-disaster city, and constructing a feature vector of the target user based on the power influence factors and the historical load data set of the target user; inputting the feature vector of the target user into a pre-constructed LSTM network model to obtain power consumption demand of the target user.

6. The method for load restoration of urban power grid based on user demand according to claim 1, characterized in that, the maintenance resources of the post-disaster city comprise maintenance team resources, path resources and traffic resources of the post-disaster city; the set of power equipment parameters comprises device parameters of normally operating power equipment and device parameters of faulty power equipment.

7. The method for load restoration of urban power system based on user demand as claimed in claim 1, wherein, the sub-model is: wherein, is the discrete variable decision result of the post-disaster recovery model of urban distribution network from the main problem, t is the time, and is the vector form of the dual variable associated with the equality and inequality constraints of continuous and discrete decision variables, q is the optimal objective function value, E , G , H is the coefficient matrix of the equality or inequality constraint containing continuous variables, is the coefficient vector corresponding to the equality or inequality constraint of continuous variables.

8. The method for load restoration of urban power system based on user demand as claimed in claim 1, wherein, the master model is: wherein, t is the time, k is the iteration number, and is the vector form of the dual variables of the post-disaster recovery resource scheduling decision variables at the time t and the k iteration, q is the optimal objective function value, is the scheduling instruction of various post-disaster recovery resources of the urban distribution network, the microgrid, and the repair team, is the coefficient vector of the discrete variable vector, A , D is the coefficient matrix of the equation or inequality constraint containing the 0 / 1 binary variable in the compact form, and is the coefficient vector corresponding to the equation or inequality constraint of the discrete variable, G is the coefficient matrix of the equation or inequality constraint containing the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint of the continuous variable, is the coefficient vector corresponding to the equation or inequality constraint satisfied by the continuous variable and the discrete variable.

9. A city grid load restoration device based on user demand, characterized by, ​ The acquisition module is configured to acquire a historical power load data set of each target user in the post-disaster city, and a set of repair resources, a power distribution network topology structure and power equipment parameters of the post-disaster city; The analysis module is configured to perform power consumption analysis on each target user based on the historical load data set of the target user, to obtain power consumption demand of the target user; The solving module is configured to input the power consumption demand of all target users, the power distribution network topology structure of the post-disaster city and the set of power equipment parameters into a pre-constructed post-disaster recovery model of a power distribution network, and solve the post-disaster recovery model with a strong repair timeliness as a target, to obtain a scheduling strategy; the post-disaster recovery model of the power distribution network includes a power system operation unit, a topology reconstruction unit and a repair resource scheduling unit; The scheduling module is configured to schedule power resources and repair resources of the post-disaster city based on the scheduling strategy; The solving module is further configured to input the power consumption demand of all target users, the power distribution network topology structure of the post-disaster city and the set of power equipment parameters into the post-disaster recovery model of the power distribution network, to obtain a mixed integer linear programming model, and construct an objective function and related constraints of the mixed integer linear programming model with a small post-disaster load shedding loss, a small micro-grid operation cost and a strong repair timeliness as targets; convert the objective function and the related constraints into a compact form, decompose the mixed integer linear programming model into a master model and a sub-model based on a Benders algorithm, and solve the master model and the sub-model, to obtain a scheduling strategy.

Images