Load recovery method and system for power distribution network containing micro-grid

Abstract

The invention discloses a micro-grid-containing power distribution network load recovery method and system, and belongs to the technical field of power distribution network fault recovery, and the method comprises the steps: taking the maximum recovery load capacity of a power distribution network as a target, considering the load condition of the power distribution network, and constructing a power distribution network load recovery model; determining micro-grid division requirements in the power distribution network, and establishing a target function of a power distribution network load recovery target model considering micro-grid division; constructing a fitness function based on the objective function, and performing micro-grid division optimization on the power distribution network by adopting an improved fuzzy adaptive simulated annealing genetic algorithm to obtain an optimal micro-grid division scheme; according to the method, the micro-grid division requirement in the power distribution network is determined, the micro-grid division mathematical model in the power distribution network is established, and the improved fuzzy adaptive simulated annealing genetic algorithm is adopted to perform micro-grid division on the power distribution network, so that the convergence of the genetic algorithm and the load recovery efficiency are improved, and the self-healing capability of the power distribution network when a fault occurs is improved; and the economic loss caused by overlong power failure time is reduced.

Description

Technical Field

[0001] This invention relates to the field of distribution network fault healing technology, and in particular to a method and system for restoring loads in a distribution network containing a microgrid. Background Technology

[0002] The power distribution network is a vital link in the power system, directly connecting end-users and playing a crucial role in ensuring healthy social development and stable lives. Traditional distribution network technologies are slow to evolve, have low levels of automation, and suffer from poor recovery capabilities after faults. Currently, with the increasing application of computer technology and the deployment of numerous automation devices in distribution networks, a safe, reliable, efficient, and comprehensive fault recovery solution is still needed to effectively improve the rapid recovery capabilities of the distribution network and ensure its safe and stable operation.

[0003] Against this backdrop, how to restore power to the distribution network by leveraging the integration of microgrids based on distributed renewable energy sources and the basic load of the distribution network, thereby achieving rapid recovery, has become a key issue in enhancing the safety and stability margin of the distribution network. Existing research considers the impact of distribution network fault types on fault recovery and obtains the final scheduling scheme through determinization of uncertain variables. However, it suffers from incomplete consideration of uncertainties and neglects load uncertainties. Traditional optimization methods such as linear programming (LP) and mixed integer programming (MILP), while rigorous in modeling, are prone to getting trapped in local optima when facing multi-constraint, nonlinear, and multi-peak problems, limiting their solution capabilities. Some scholars have used particle swarm optimization and gray wolf algorithms for modeling and solving, but these algorithms still suffer from problems such as premature convergence, insufficient convergence accuracy, or sensitivity to parameters, making it difficult to simultaneously meet the requirements of solution efficiency and computational accuracy.

[0004] In summary, there is an urgent need for a load restoration method for distribution networks containing microgrids to enhance the stability margin of distribution network security. Summary of the Invention

[0005] This invention provides a method and system for restoring the load of a distribution network containing a microgrid, which solves the problem of the distribution network being difficult to restore after a fault and effectively enhances the stability margin of the distribution network safety.

[0006] In view of this, the first aspect of the present invention provides a method for load restoration of a distribution network containing a microgrid, the method comprising:

[0007] With the goal of maximizing the load restoration capacity of the distribution network, and taking into account the load conditions of the distribution network, a distribution network load restoration model is constructed.

[0008] Define the requirements for microgrid partitioning in the distribution network, and establish the objective function of the distribution network load recovery target model that takes microgrid partitioning into account;

[0009] A fitness function is constructed based on the objective function, and an improved fuzzy adaptive simulated annealing genetic algorithm is used to optimize the microgrid partitioning of the distribution network and obtain the optimal microgrid partitioning scheme.

[0010] Optionally, the expression for constructing the distribution network load recovery model is:

[0011] ;

[0012] In the formula, This indicates the maximum value of the load restored in the distribution network. The number of loads restored to the distribution network. The weight value P represents the importance of the distribution network load. Li Indicates the first The active power of the load, P SLi It represents the sum of the active power of all loads at the same level.

[0013] Optionally, the load recovery model of the distribution network is established with constraints including power balance constraints, output constraints, node voltage magnitude constraints, and line power flow constraints.

[0014] Optionally, the requirements for defining the microgrid division in the distribution network include: the maximum total restored load, the minimum number of microgrids, each microgrid containing at least one microgrid or black start unit, and the microgrid area not containing only distributed power sources.

[0015] Optionally, based on the requirements for microgrid partitioning in the distribution network, the objective function of the load recovery target model of the distribution network considering microgrid partitioning is established as follows:

[0016] ;

[0017] In the formula, This represents the maximum value of the restored load in the distribution network considering microgrid partitioning. K represents the number of nodes in the microgrid, and λ1 and λ2 represent the weighting coefficients for the influence of load restoration and the number of switch disconnections on the objective function value, respectively. i Indicates whether the load is included in the microgrid, P Li Indicates the first The active power of the load, P SLi N represents the sum of the active power of all loads within the same load class. s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

[0018] Optionally, the fitness function constructed based on the objective function is as follows:

[0019] In each microgrid partitioning scheme, the load restoration amount of the microgrid The expression is:

[0020] ;

[0021] Each microgrid contains Nodes, if containing For a microgrid, the fitness function for the distribution network load recovery is established as follows:

[0022] ;

[0023] In the formula, This represents the maximum value of the distribution network load restoration under the fitness function. λ1 and λ2 represent the weighting coefficients of the influence of load restoration and the number of switch disconnections on the fitness function value, respectively. Indicates the first Load restoration capacity of a microgrid under a microgrid partitioning scheme, N s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

[0024] Optionally, the process of using the improved fuzzy adaptive simulated annealing genetic algorithm to optimize the microgrid partitioning of the distribution network is as follows:

[0025] Based on the selection mechanism of the fuzzy adaptive simulated annealing genetic algorithm, the generated initial set of microgrid partitioning schemes is selected and eliminated. The calculated proportion of the fitness value of a single microgrid partitioning scheme in the set is used as the selection probability of that individual microgrid partitioning scheme. Individual microgrid partitioning schemes are retained or eliminated according to the roulette wheel selection rule. The selection probability of a single microgrid partitioning scheme is... The calculation formula is:

[0026] ;

[0027] In the formula, This represents the probability of selecting a single microgrid partitioning scheme. This represents the fitness of a single microgrid partitioning scheme within the set of microgrid partitioning schemes. This represents the number of individual microgrid partitioning schemes (individuals) within the set of microgrid partitioning schemes.

[0028] Alternatively, the selection mechanism can be improved as follows:

[0029] The fitness value of a single microgrid partitioning scheme is transformed based on a genetic process, thereby changing the selection probability of the single microgrid partitioning scheme. The adaptive selection probability of a single microgrid partitioning scheme is calculated as follows:

[0030] ;

[0031] f '( x i The fitness of a single microgrid partitioning scheme is transformed, and the calculation formula is as follows:

[0032] ;

[0033] In the formula, f represents the adaptive selection probability of a single microgrid partitioning scheme. max and f min These represent the maximum fitness value and minimum fitness value of a single microgrid partitioning scheme in the current set of microgrid partitioning schemes, respectively, where k is the current generation number, and K is the minimum fitness value. max Let α be the maximum number of generations, and β be constants greater than zero.

[0034] Optionally, a single microgrid partitioning scheme is represented by binary codes of 1s and 0s, where each bit of the code corresponds to a load node, with 1 indicating that the node is included in the microgrid and 0 indicating that the node is not included in the microgrid.

[0035] The microgrid partitioning process begins with initializing the set of microgrid partitioning schemes, randomly generating several binary-coded microgrid partitioning schemes, and prioritizing the retention of the microgrid partitioning scheme with the highest fitness function value based on the fitness function value corresponding to each microgrid partitioning scheme. Subsequently, a cross operation is performed on the microgrid partitioning schemes to select the two best-performing microgrid partitioning schemes.

[0036] The crossover and mutation mechanisms of the fuzzy adaptive simulated annealing genetic algorithm are improved, and the adaptive expressions for the crossover and mutation probabilities of the microgrid partitioning scheme are obtained as follows:

[0037] ;

[0038] In the formula, and The improved legacy algorithm obtains the crossover and mutation probabilities of microgrid partitioning schemes, f, respectively. avg Let f' be the average fitness value of the set of microgrid partitioning schemes, f' be the larger fitness value among the two microgrid partitioning schemes to be crossed, and f be the fitness value of the microgrid partitioning scheme to be mutated; p c1 p c2 and p m1 p m2 These represent the crossover and mutation probabilities of the microgrid partitioning schemes in the original legacy algorithm;

[0039] The optimal selection of microgrid partitioning schemes from the set of cross- and mutated offspring microgrid partitioning schemes is constituted by a fuzzy adaptive cross- or mutated annealing operation, as shown in the following expression:

[0040] ;

[0041] In the formula, and It is the acceptance probability under the simulated annealing mechanism, f x1 f x2 f xi and f xj These are the fitness values ​​for the corresponding microgrid partitioning schemes, and T is the simulated annealing temperature;

[0042] The fuzzy adaptive crossover or mutation annealing operation is as follows:

[0043] The microgrid partitioning scheme generates a new microgrid partitioning scheme x through fuzzy adaptive mutation operation. k If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is greater than that of the old microgrid partitioning scheme, then the new microgrid partitioning scheme x will be used. k Replaces the old microgrid partitioning scheme;

[0044] If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is less than that of the old scheme, then the Boltiziman mechanism is used to accept the new microgrid partitioning scheme. The acceptance probability of the new microgrid partitioning scheme under the Boltiziman mechanism is... The formula is as follows:

[0045] ;

[0046] In the formula, f x1 and f xk These are the corresponding microgrid partitioning schemes. and the new microgrid partitioning scheme x k fitness value;

[0047] Parameter settings: Initial simulated annealing temperature set to T0, cooling parameter set to T. k ;

[0048] Calculate the algebra k = 1, and set the initial temperature to T = T0;

[0049] Calculate the fitness value of the microgrid partitioning scheme, and perform adaptive selection and fuzzy adaptive cross-annealing operations on the microgrid partitioning schemes in the set of microgrid partitioning schemes in turn;

[0050] Increasing the algebraic value k = k + 1, the temperature decreases by T = T × (T k ) k Then return to the operation step of calculating the fitness value of the microgrid partitioning scheme again until the number of iterations ends, and output the optimal microgrid partitioning scheme.

[0051] A second aspect of the present invention provides a distribution network load restoration system including a microgrid, the system comprising:

[0052] The construction unit aims to maximize the load restoration of the distribution network and considers the load conditions of the distribution network to construct a distribution network load restoration model;

[0053] The improved unit enhances the fuzzy adaptive simulated annealing genetic algorithm by incorporating adaptive and simulated annealing mechanisms, thereby improving the algorithm's local search capability.

[0054] The computing unit uses an improved fuzzy adaptive simulated annealing genetic algorithm to optimize the microgrid partitioning scheme of the distribution network, obtain the optimal microgrid partitioning scheme, and obtain the recovery status of related loads after the fault encountered by the distribution network under various operating constraints.

[0055] As can be seen from the above technical solutions, the present invention has the following advantages:

[0056] This invention constructs a distribution network load recovery model with the goal of maximizing the load recovery capacity of the distribution network, taking into account the power generation capacity of microgrids and the load conditions of the distribution network. It utilizes operational constraints such as distribution network power balance, generator output limitations, distribution network load types, line power flow, and node voltage to effectively improve the maximum recovery capacity of the distribution network under fault conditions, thus enhancing the safety and stability margin of the distribution network. Furthermore, by clarifying the requirements for microgrid partitioning within the distribution network and establishing a mathematical model for microgrid partitioning, an improved fuzzy adaptive simulated annealing genetic algorithm is used to partition the distribution network into microgrids. This avoids premature convergence in genetic algorithms, which can lead to slow convergence or getting trapped in local optima, and improves local search capabilities. This allows for the determination of distribution network load recovery under various operational constraints, providing technical support for the planning and scheduling strategies of distribution network fault self-healing. Attached Figure Description

[0057] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art 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.

[0058] Figure 1 A schematic flowchart of a load restoration method for a distribution network containing a microgrid, provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the microgrid partitioning process based on the fuzzy adaptive simulated annealing genetic algorithm provided in an embodiment of the present invention; Figure 3 This is a schematic diagram of the framework of a distribution network load restoration system containing a microgrid, provided in an embodiment of the present invention. Detailed Implementation

[0059] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0060] Please see Figure 1 This invention provides a method for restoring loads in a distribution network containing a microgrid, comprising:

[0061] With the goal of maximizing the load recovery of the distribution network, and taking into account the power generation capacity of the microgrid and the load conditions of the distribution network, a distribution network load recovery model is constructed based on operational constraints such as distribution network power balance, generator output limits, distribution network load types, line power flow, and node voltage.

[0062] It should be noted that the distribution network load conditions obtained under various operating constraints directly affect the safety and stability of the distribution network. On the one hand, the integration of new energy sources into the system lacks the rotational inertia and regulation capabilities of traditional synchronous generators, leading to a decrease in overall system inertia and a deterioration in voltage stability. On the other hand, the types and capacities of loads integrated into the system are accumulating. To address the issue of distribution network load recovery after a fault, the expression for the distribution network load recovery model is constructed as follows:

[0063] ;

[0064] In the formula, This indicates the maximum value of the load restored in the distribution network. The number of loads restored to the distribution network. The weight value P represents the importance of the distribution network load. Li Indicates the first The active power of the load, P SLi It represents the sum of the active power of all loads at the same level.

[0065] It should be noted that in the distribution network load recovery model, the power system is assumed to contain... Each microgrid node A conventional generator, For each new energy power station, the following constraints are designed.

[0066] The load recovery model for the distribution network is established with constraints including power balance constraints, output constraints, node voltage magnitude constraints, and line power flow constraints. Specifically:

[0067] (1) Power balance constraint:

[0068] For each node The active power balance constraint must be satisfied:

[0069] ;

[0070] In the formula, This refers to the system load after coordinated regulation of distributed renewable energy and adjustable load. , Node voltage; , Let be the real and imaginary parts of the nodal admittance; This is the phase angle difference.

[0071] Output constraints include conventional generator output constraints and renewable energy output constraints.

[0072] (2) Conventional generator output constraints:

[0073] ;

[0074] In the formula, , These represent the minimum and maximum output values ​​of a conventional generator. This is the rated output value of a conventional generator.

[0075] (3) Constraints on new energy output:

[0076] ;

[0077] In the formula, To achieve maximum output of photovoltaic and wind power under current weather conditions, and The outputs are solar and wind power, respectively. This indicates the output of photovoltaic and wind power under the current meteorological conditions.

[0078] (4) Node voltage amplitude constraints:

[0079] ;

[0080] In the formula, , Let be the voltage limit at node i. Let be the voltage value at node i.

[0081] (5) Power flow constraints of the line:

[0082] For any transmission line of the distribution network ,satisfy:

[0083] ;

[0084] In the formula, For the node To the node Complex power flow; This represents the upper limit of the line's transmission capacity.

[0085] Define the requirements for microgrid division in the distribution network and establish the objective function of the distribution network load recovery target model that takes into account microgrid division.

[0086] The requirements for defining microgrids in a distribution network include:

[0087] The total amount of load restored is the largest;

[0088] The number of microgrids is minimized (the number of system disconnection points is minimized);

[0089] Each microgrid contains at least one microgrid or black start unit to regulate the voltage and frequency of the microgrid area;

[0090] Considering the operating characteristics of distributed power sources and the requirements of the black start process of the distribution network, it is required that the microgrid area cannot contain only distributed power sources.

[0091] When dividing a distribution network into microgrids, the optimal microgrid division scheme is found by simultaneously maximizing the load recovery and minimizing the number of switching operations, using a weighted balance between these two objectives. Under the premise of satisfying the distribution network operation constraints, the maximum load recovery amount is obtained. The objective function is the final objective function obtained by taking into account the microgrid division while maximizing the load recovery objective function. The microgrid division result and the recoverable load under fault conditions are determined by the switch-off status in the distribution network.

[0092] Therefore, based on the above requirements for microgrid partitioning in the distribution network, the objective function of the distribution network load restoration target model considering microgrid partitioning is as follows:

[0093]

[0094] In the formula, This represents the maximum value of the restored load in the distribution network considering microgrid partitioning. K represents the number of nodes in the microgrid, and λ1 and λ2 represent the weighting coefficients for the influence of load restoration and the number of switch disconnections on the objective function value, respectively. i Indicates whether the load is included in the microgrid, P Li Indicates the first The active power of the load, P SLi N represents the sum of the active power of all loads within the same load class. s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

[0095] The optimal solution process of an algorithm typically does not rely on external information, but only calculates the optimal solution based on the fitness function. Therefore, the construction of the fitness function directly affects the optimization process and convergence of the algorithm. The fitness value is a numerical value that measures the quality of the algorithm's iterative calculation results. When the fitness value is optimal, it indicates that the algorithm's iterative result is optimal. The fitness value is inherent in the algorithm itself, but the calculation of the fitness value varies depending on different objective functions and scenarios. To facilitate the calculation of the fitness value, the objective function is transformed into a fitness function.

[0096] The fitness function is constructed based on the objective function as follows:

[0097] In each microgrid partitioning scheme, the load restoration amount of the microgrid The expression is:

[0098] ;

[0099] Each microgrid contains Nodes, if containing For a microgrid, the fitness function for the distribution network load recovery is established as follows:

[0100] ;

[0101] In the formula, This represents the maximum value of the distribution network load restoration under the fitness function. λ1 and λ2 represent the weighting coefficients of the influence of load restoration and the number of switch disconnections on the fitness function value, respectively. Indicates the first Load restoration capacity of a microgrid under a microgrid partitioning scheme, N s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

[0102] To improve the power supply reliability, operational flexibility, and energy utilization efficiency of the distribution network, the application of genetic algorithms in the microgrid partitioning of the distribution network requires a one-to-one correspondence between the microgrid partitioning schemes and the elements of the algorithm, and iterative optimization through evolutionary operations. First, the core elements of the algorithm are determined: the population corresponds to multiple sets of different microgrid partitioning schemes, and the individual is a single partitioning scheme, represented by binary encoding (for example, each bit of the encoding corresponds to a load node, "1" indicates that the node is included in the microgrid, and "0" indicates that it is not included; the encoding also implicitly contains information about the number of times the switch has been opened).

[0103] The specific partitioning process begins with initializing the population: several individuals with binary codes (i.e., different combinations of load inclusion) are randomly generated, each corresponding to a complete set of microgrid partitioning candidate schemes. Next, a selection operation is used to screen for superior schemes: the fitness function value corresponding to each individual is calculated (considering both load recovery ratio and switching operation penalty), prioritizing the retention of individuals with higher fitness function values ​​to ensure that better partitioning schemes proceed to the next round of evolution. Then, a crossover operation is performed: two well-performing individuals are selected, and parts of their codes are swapped (e.g., the inclusion or exclusion status of certain load nodes is exchanged), generating a new microgrid partitioning scheme that combines the advantages of both. To avoid getting trapped in local optima, a mutation operation is performed: a few bits in the codes of some individuals are randomly modified (e.g., changing an excluded load node to an included one), introducing the possibility of new microgrid partitioning schemes.

[0104] Through multiple rounds of selection, crossover, and mutation iterations, individuals (microgrid partitioning schemes) in the population will gradually evolve towards maximizing the fitness function. The optimal individual (microgrid partitioning scheme) obtained in the end is the optimal microgrid partitioning scheme that balances maximizing load recovery and minimizing switching operations.

[0105] A fitness function is constructed based on the objective function, and an improved fuzzy adaptive simulated annealing genetic algorithm is used to optimize the microgrid partitioning of the distribution network and obtain the optimal microgrid partitioning scheme.

[0106] First, the fuzzy adaptive simulated annealing genetic algorithm is improved.

[0107] Based on the selection mechanism of the fuzzy adaptive simulated annealing genetic algorithm, the generated initial (population) set of microgrid partitioning schemes is selected and eliminated. The proportion of the fitness value of a single microgrid partitioning scheme (individual) in the set (population) is used as the selection probability of a single microgrid partitioning scheme (individual). Individual microgrid partitioning schemes (individuals) are retained or eliminated according to the rules of roulette wheel selection. The selection probability of a single microgrid partitioning scheme... The calculation formula is:

[0108] ;

[0109] In the formula, This represents the probability of selecting a single microgrid partitioning scheme. This represents the fitness of a single microgrid partitioning scheme (individual) within the set (population) of microgrid partitioning schemes. This represents the number of individual microgrid partition schemes (individuals) in the set (population) of microgrid partition schemes.

[0110] When the fitness value of a microgrid partition scheme (individual) is high, its proportion is large, and its probability of being selected is also high. This directly leads to a rapid increase in the proportion of excellent microgrid partition schemes (individuals), resulting in similar fitness values ​​among individuals entering the crossover phase. In some cases, individuals of this type may even dominate the entire population, forming super-individuals that mask the selection process in subsequent genetics. This selection mechanism disrupts population diversity during the selection phase after the initial population is generated, making the genetic algorithm highly susceptible to getting trapped in local optima, resulting in premature convergence, or worsening the algorithm's search ability and reducing convergence.

[0111] Crossover and mutation mechanisms: Crossover and mutation operations significantly improve population diversity, expand the search space of the genetic algorithm, enhance its search capability, and ensure its effectiveness. Crossover and mutation operations are typically performed based on given probabilities, and these probabilities are crucial to the performance of the genetic algorithm, determining its search capability and convergence.

[0112] The selection mechanism is improved as follows:

[0113] The fitness value of a single microgrid partitioning scheme is transformed based on a genetic process, thereby changing the selection probability of the single microgrid partitioning scheme. The adaptive selection probability of a single microgrid partitioning scheme is calculated as follows:

[0114] ;

[0115] f '( x i The fitness of a single microgrid partitioning scheme is transformed, and the calculation formula is as follows:

[0116] ;

[0117] In the formula, f represents the adaptive selection probability of a single microgrid partitioning scheme. max and f min These represent the maximum fitness value and minimum fitness value of a single microgrid partitioning scheme in the current set of microgrid partitioning schemes, respectively, where k is the current generation number, and K is the minimum fitness value. max Let α be the maximum number of generations, and β be constants greater than zero.

[0118] In the early stages of inheritance, let:

[0119] ;

[0120] In the later stages of inheritance, let:

[0121] ;

[0122] At this point, the probability of selecting the optimal microgrid partitioning scheme increases, improving the sensitivity of the selection and accelerating the convergence of the algorithm.

[0123] The crossover and mutation mechanisms of the fuzzy adaptive simulated annealing genetic algorithm are improved, and the adaptive expressions for the crossover and mutation probabilities of the microgrid partitioning scheme are obtained as follows:

[0124] ;

[0125] In the formula, and The improved legacy algorithm obtains the crossover and mutation probabilities of microgrid partitioning schemes, f, respectively. avg Let f' be the average fitness value of the set of microgrid partitioning schemes, f' be the larger fitness value among the two microgrid partitioning schemes to be crossed, and f be the fitness value of the microgrid partitioning scheme to be mutated; p c1 p c2 and p m1 p m2 These represent the crossover and mutation probabilities of the microgrid partitioning schemes in the original legacy algorithm;

[0126] The optimal selection of microgrid partitioning schemes from the set of cross- and mutated offspring microgrid partitioning schemes is constituted by a fuzzy adaptive cross- or mutated annealing operation, as shown in the following expression:

[0127] ;

[0128] In the formula, and It is the acceptance probability under the simulated annealing mechanism, f x1 f x2 f xi and f xj These are the fitness values ​​for the corresponding microgrid partitioning schemes, and T is the simulated annealing temperature;

[0129] The fuzzy adaptive crossover or mutation annealing operation is as follows:

[0130] The microgrid partitioning scheme generates a new microgrid partitioning scheme x through fuzzy adaptive mutation operation. k If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is greater than that of the old microgrid partitioning scheme, then the new microgrid partitioning scheme x will be used. k Replaces the old microgrid partitioning scheme;

[0131] If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is less than that of the old scheme, then the Boltiziman mechanism is used to accept the new microgrid partitioning scheme, i.e.: P kWhen the value is greater than or equal to 0,1, the new microgrid partitioning scheme x is used. k Alternatives must be accepted; otherwise, the new microgrid partitioning scheme will not be accepted. k Acceptance probability of new microgrid partitioning schemes under the Boltiziman mechanism The formula is as follows:

[0132] ;

[0133] In the formula, f x1 and f xk These are the corresponding microgrid partitioning schemes. and the new microgrid partitioning scheme x k fitness value;

[0134] like Figure 2 As shown, the distribution network parameters and node weighting information are generated first.

[0135] Algorithm parameter initialization settings, generation of the first generation population, simulated annealing initial temperature set to T0, cooling parameter set to T k ;

[0136] Calculate the algebra k = 1, and set the initial temperature to T = T0;

[0137] Decode the individual population, calculate the fitness value of the microgrid partitioning scheme, and perform adaptive selection, optimal retention strategy, fuzzy adaptive mutation annealing, and fuzzy adaptive cross annealing operations on the microgrid partitioning schemes in the microgrid partitioning scheme set in sequence.

[0138] If the number of iterations has not been reached, increase the algebra k = k + 1, and decrease the temperature T = T×(T k ) k The algorithm returns to the step of calculating the fitness value of the microgrid partitioning scheme again until the number of iterations ends. Then, it selects the individual with the highest fitness in the population, decodes it to obtain the final microgrid partitioning result, that is, the scheme that maximizes the load recovery of the distribution network in the microgrid partitioning, outputs the optimal microgrid partitioning scheme (the microgrid partitioning scheme that takes into account both maximizing load recovery and minimizing switching operations), and the algorithm ends.

[0139] If the load or power output of the distribution network fluctuates significantly, data can be updated periodically, new schemes can be added to the population, and the above process can be repeated to achieve dynamic optimization.

[0140] Reference Figure 3 The present invention provides a distribution network load restoration system containing a microgrid, the system comprising:

[0141] Building unit 201, with the goal of maximizing the load restoration of the distribution network, and considering the load conditions of the distribution network, constructs a distribution network load restoration model;

[0142] Improved unit 202 improves the fuzzy adaptive simulated annealing genetic algorithm by adding an adaptive mechanism and a simulated annealing mechanism;

[0143] The computing unit 203 uses an improved fuzzy adaptive simulated annealing genetic algorithm to find the optimal microgrid partitioning scheme for the distribution network.

[0144] This invention provides a method and system for load restoration in distribution networks containing microgrids. By constructing a multi-constraint nonlinear optimization model with the objective function of maximizing the load restoration amount of the distribution network, it fully considers the key constraints in power system operation, effectively improving the load restoration problem in the event of a distribution network fault, and enhancing the safety and stability margin of the distribution network power supply. In this method, firstly, the power generation capacity of the microgrid and the load situation of the distribution network are comprehensively considered, taking into account operational constraints such as distribution network power balance, generator output limits, distribution network load types, line power flow, and node voltage. Secondly, by clarifying the requirements for microgrid partitioning in the distribution network, a mathematical model for microgrid partitioning in the distribution network is established. An improved fuzzy adaptive simulated annealing genetic algorithm is used to partition the distribution network into microgrids, avoiding premature convergence of the genetic algorithm, which can lead to slow convergence or getting trapped in local optima, thus improving the local search capability and obtaining the load restoration status of the distribution network. This provides technical support for the formulation of power grid safety and stability operation strategies.

[0145] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0146] In the embodiments provided by this invention, it should be understood that the disclosed methods and systems can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0147] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0148] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0149] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this 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. This computer software product is stored in a storage medium and includes several instructions to cause a computer device to execute all or part of the steps of the methods of the various embodiments of the present invention.

[0150] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for load restoration in a distribution network containing a microgrid, characterized in that, include: With the goal of maximizing the load restoration capacity of the distribution network, and taking into account the load conditions of the distribution network, a distribution network load restoration model is constructed. Define the requirements for microgrid partitioning in the distribution network, and establish the objective function of the distribution network load recovery target model that takes microgrid partitioning into account; A fitness function is constructed based on the objective function, and an improved fuzzy adaptive simulated annealing genetic algorithm is used to optimize the microgrid partitioning of the distribution network and obtain the optimal microgrid partitioning scheme.

2. The method for restoring loads in a distribution network containing a microgrid according to claim 1, characterized in that, The expression for constructing the distribution network load recovery model is as follows: ； In the formula, This indicates the maximum value of the load restored in the distribution network. The number of loads restored to the distribution network. The weight value P represents the importance of the distribution network load. Li Indicates the first The active power of a load, P SLi It represents the sum of the active power of all loads at the same level.

3. The method for restoring loads in a distribution network containing a microgrid according to claim 2, characterized in that, The load recovery model of the distribution network is established with constraints including power balance constraints, output constraints, node voltage magnitude constraints, and line power flow constraints.

4. The method for restoring loads in a distribution network containing a microgrid according to claim 1, characterized in that, The requirements for defining the microgrid division in the distribution network include: the maximum total restored load, the minimum number of microgrids, each microgrid containing at least one microgrid or black start unit, and the microgrid area not containing only distributed power sources.

5. The method for restoring loads in a distribution network containing a microgrid according to claim 4, characterized in that, Based on the requirements for microgrid partitioning in the distribution network, the objective function of the load recovery target model for the distribution network considering microgrid partitioning is established as follows: ； In the formula, This represents the maximum value of the restored load in the distribution network considering microgrid partitioning. K represents the number of nodes in the microgrid, and λ1 and λ2 represent the weighting coefficients for the influence of load restoration and the number of switch disconnections on the objective function value, respectively. i Indicates whether the load is included in the microgrid, P Li Indicates the first The active power of a load, P SLi N represents the sum of the active power of all loads within the same load class. s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

6. The method for restoring loads in a distribution network containing a microgrid according to claim 1, characterized in that, The fitness function constructed based on the objective function is as follows: In each microgrid partitioning scheme, the load restoration amount of the microgrid The expression is: ； Each microgrid contains Nodes, if containing For a microgrid, the fitness function for the distribution network load recovery is established as follows: ； In the formula, This represents the maximum value of the distribution network load restoration under the fitness function. λ1 and λ2 represent the weighting coefficients of the influence of load restoration and the number of switch disconnections on the fitness function value, respectively. Indicates the first Load restoration capacity of a microgrid under a microgrid partitioning scheme, N s This represents the number of switches disconnected in the distribution network when dividing the microgrid.

7. The method for load restoration of a distribution network containing a microgrid according to claim 1, characterized in that, The process of using the improved fuzzy adaptive simulated annealing genetic algorithm to optimize the microgrid partitioning of the distribution network is as follows: Based on the selection mechanism of the fuzzy adaptive simulated annealing genetic algorithm, the generated initial set of microgrid partitioning schemes is selected and eliminated. The calculated proportion of the fitness value of a single microgrid partitioning scheme in the set is used as the selection probability of that individual microgrid partitioning scheme. Individual microgrid partitioning schemes are retained or eliminated according to the roulette wheel selection rule. The selection probability of a single microgrid partitioning scheme is... The calculation formula is: ； In the formula, This represents the probability of selecting a single microgrid partitioning scheme. This represents the fitness of a single microgrid partitioning scheme within the set of microgrid partitioning schemes. This represents the number of individual microgrid partitioning schemes in the set of microgrid partitioning schemes.

8. The method for restoring loads in a distribution network containing a microgrid according to claim 7, characterized in that: The improvement to the selection mechanism is as follows: The fitness value of a single microgrid partitioning scheme is transformed based on a genetic process, thereby changing the selection probability of the single microgrid partitioning scheme. The adaptive selection probability of a single microgrid partitioning scheme is calculated as follows: ； f '( x i The fitness of a single microgrid partitioning scheme is transformed, and the calculation formula is as follows: ； In the formula, f represents the adaptive selection probability of a single microgrid partitioning scheme. max and f min These represent the maximum fitness value and minimum fitness value of a single microgrid partitioning scheme in the current set of microgrid partitioning schemes, respectively, where k is the current generation number, and K is the minimum fitness value. max Let α be the maximum number of generations, and β be constants greater than zero.

9. A method for restoring loads in a distribution network containing a microgrid according to claim 8, characterized in that: A single microgrid partitioning scheme is represented by binary codes of 1s and 0s. Each bit of the code corresponds to a load node, with 1 indicating that the node is included in the microgrid and 0 indicating that the node is not included in the microgrid. The microgrid partitioning process begins with initializing the set of microgrid partitioning schemes, randomly generating several binary-coded microgrid partitioning schemes, and prioritizing the retention of the microgrid partitioning scheme with the highest fitness function value based on the fitness function value corresponding to each microgrid partitioning scheme. Subsequently, a cross operation is performed on the microgrid partitioning schemes to select the two best-performing microgrid partitioning schemes. The crossover and mutation mechanisms of the fuzzy adaptive simulated annealing genetic algorithm are improved, and the adaptive expressions for the crossover and mutation probabilities of the microgrid partitioning scheme are obtained as follows: ； In the formula, and The improved legacy algorithm obtains the crossover and mutation probabilities of microgrid partitioning schemes, f, respectively. avg Let f' be the average fitness value of the set of microgrid partitioning schemes, f' be the larger fitness value among the two microgrid partitioning schemes to be crossed, and f be the fitness value of the microgrid partitioning scheme to be mutated; p c1 p c2 and p m1 p m2 These represent the crossover and mutation probabilities of the microgrid partitioning schemes in the original legacy algorithm; The optimal selection of microgrid partitioning schemes from the set of cross- and mutated offspring microgrid partitioning schemes is constituted by a fuzzy adaptive cross- or mutated annealing operation, as shown in the following expression: ； In the formula, and It is the acceptance probability under the simulated annealing mechanism, f x1 f x2 f xi and f xj These are the fitness values ​​for the corresponding microgrid partitioning schemes, and T is the simulated annealing temperature; The fuzzy adaptive crossover or mutation annealing operation is as follows: The microgrid partitioning scheme generates a new microgrid partitioning scheme x through fuzzy adaptive mutation operation. k If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is greater than that of the old microgrid partitioning scheme, then the new microgrid partitioning scheme x will be used. k Replaces the old microgrid partitioning scheme; If the new microgrid partitioning scheme x k If the fitness of the new microgrid partitioning scheme is less than that of the old scheme, then the Boltiziman mechanism is used to accept the new microgrid partitioning scheme. The acceptance probability of the new microgrid partitioning scheme under the Boltiziman mechanism is... The formula is as follows: ； In the formula, f x1 and f xk These are the corresponding microgrid partitioning schemes. and the new microgrid partitioning scheme x k fitness value; Parameter settings: Initial simulated annealing temperature set to T0, cooling parameter set to T. k ; Calculate the algebra k = 1, and set the initial temperature to T = T0; Calculate the fitness value of the microgrid partitioning scheme, and perform adaptive selection and fuzzy adaptive cross-annealing operations on the microgrid partitioning schemes in the set of microgrid partitioning schemes in turn; Increasing the algebraic value k = k + 1, the temperature decreases by T = T × (T k ) k Then return to the operation step of calculating the fitness value of the microgrid partitioning scheme again until the number of iterations ends, and output the optimal microgrid partitioning scheme.

10. A load restoration system for a distribution network containing a microgrid, characterized in that, include: The construction unit aims to maximize the load restoration of the distribution network and considers the load conditions of the distribution network to construct a distribution network load restoration model; The improved unit enhances the fuzzy adaptive simulated annealing genetic algorithm by incorporating adaptive and simulated annealing mechanisms. The computing unit uses an improved fuzzy adaptive simulated annealing genetic algorithm to find the optimal microgrid partitioning scheme for the distribution network.

Images