Method for evaluating new energy carrying capacity of distribution network based on manta ray foraging optimization algorithm

Abstract

The application provides a power distribution network new energy carrying capacity evaluation method based on a multi-strategy improved adaptive manta ray foraging optimization algorithm (MSAMRFO), which comprises the following steps: inputting power distribution network framework data to be calculated for carrying capacity, and determining a new energy access node according to resource endowment of the power distribution network; forming a power distribution network new energy carrying capacity index evaluation system, constructing a carrying capacity evaluation calculation model of new energy access; and solving the established carrying capacity evaluation calculation model by using the MSAMRFO algorithm. The index constraints considered in the model established by the application are more comprehensive. Meanwhile, the application adopts the multi-strategy improved adaptive manta ray foraging optimization algorithm, which is a more efficient and accurate solving method.

Description

Technical Field

[0001] This invention relates to the field of renewable energy carrying capacity assessment technology for distribution networks, and in particular to a method for assessing renewable energy carrying capacity for distribution networks based on a manta ray foraging optimization algorithm. Background Technology

[0002] my country has a vast territory and abundant renewable energy sources such as wind and solar power. To promote the construction of a new power system based on new energy sources and help my country achieve its "dual-carbon goals," the country is vigorously promoting the construction of new energy sources and large-scale grid connection of renewable energy. On June 20, 2021, according to the "Notice on Submitting Pilot Programs for Rooftop Distributed Photovoltaic Development in Entire Counties (Cities, Districts)," the pilot program for promoting rooftop distributed photovoltaic development in entire counties (cities, districts) was launched. The main purpose is to fully mobilize and leverage local enthusiasm and expand the scale of rooftop distributed photovoltaic construction. The "Notice of the National Energy Administration on Relevant Matters Concerning the Development and Construction of Wind Power and Photovoltaic Power Generation in 2021" encourages large-scale grid connection of new energy sources.

[0003] However, due to the volatility and randomness of renewable energy output, large-scale grid connection of renewable energy not only alters power flow distribution but also seriously threatens the power quality of the distribution network, causing problems such as voltage deviation, voltage fluctuation, and failure to pass thermal stability verification, thus affecting the safe and stable operation of the power system. Therefore, it is essential to assess the renewable energy carrying capacity of the distribution network before planning and constructing renewable energy sources.

[0004] The carrying capacity of renewable energy in a distribution network refers to the maximum capacity limit that distributed generation can be connected to the grid without violating the constraints of the system's normal operating technical indicators. Many factors affect the grid connection capacity of distributed generation, such as voltage deviation, voltage fluctuation, thermal stability verification, and short-circuit current.

[0005] Existing technologies include the following analyses of factors affecting the grid connection capacity of distributed generation sources:

[0006] The paper "Distributed Generation Access Capacity Analysis Considering Voltage Quality and Short-Circuit Capacity Constraints," published in "Power System Technology" in 2016 by Zou Hongliang, Han Xiangyu, Liao Qingfen, Liu Dichun, Zhu Zhenshan, Chen Wei, and others, considered voltage quality and short-circuit capacity constraints and used a genetic algorithm to evaluate the renewable energy carrying capacity of the distribution network. However, this scheme only considered voltage quality and short-circuit capacity in the carrying capacity assessment, neglecting thermal stability constraints. Furthermore, using a genetic algorithm is prone to getting trapped in local optima and requires a large number of iterations, resulting in low efficiency.

[0007] Furthermore, traditional methods for calculating the carrying capacity of distribution networks often employ a segmented calculation approach. This method divides the renewable energy capacity at different nodes into a series of discrete capacity values ​​with a certain step size ΔS. The renewable energy capacity changes segment by segment according to the step size ΔS until any one of the indicators in the carrying capacity assessment exceeds its limit. The accuracy of this method largely depends on the size of the step size ΔS. If the step size is not chosen appropriately, the reliability of the carrying capacity assessment results will be low. At the same time, this method has low solution efficiency and is not suitable for large-scale, multi-node renewable energy integration.

[0008] In summary, the constraints considered in the traditional genetic algorithm-based assessment model for the carrying capacity of new energy sources in distribution networks, which takes into account voltage quality and short-circuit current, are not comprehensive, and the solution algorithm cannot guarantee global optimality and computational efficiency. Summary of the Invention

[0009] This invention proposes a new energy carrying capacity assessment method for distribution networks based on the Multi-Strategy Improved Adaptive Manta Ray Foraging Optimization (MSAMRFO) algorithm. This method addresses the problems in existing technologies where the assessment models do not comprehensively consider constraints, and the solution algorithms cannot guarantee global optimality and computational efficiency.

[0010] This invention employs a heuristic algorithm with faster convergence speed and stronger global optimization capability—the Multi-Strategy Improved Adaptive Manta Ray Foraging Optimization Algorithm (MSAMRFO)—to calculate the carrying capacity of a distribution network for new energy sources. The Manta Ray Foraging Optimization Algorithm (MRFO) mimics the foraging process of manta rays in the ocean, mathematically modeling different predation strategies and mathematically describing the way individual manta rays update their positions. MRFO has three predation strategies: chain foraging, spiral foraging, and tumbling foraging.

[0011] The objective of this invention is achieved through the following technical solution:

[0012] The method for assessing the renewable energy carrying capacity of distribution networks based on the manta ray foraging optimization algorithm includes the following steps:

[0013] Input the distribution network structure data for which the carrying capacity needs to be calculated, and determine the access nodes for new energy sources based on the resource endowment of the distribution network;

[0014] Establish a new energy carrying capacity index evaluation system for distribution networks and construct a carrying capacity evaluation calculation model for new energy access;

[0015] The MSAMRFO algorithm was used to solve the established bearing capacity assessment calculation model.

[0016] Furthermore, the aforementioned system for evaluating the carrying capacity of new energy in the distribution network, and the construction of a calculation model for evaluating the carrying capacity of new energy access, include:

[0017] Define the boundary conditions for the bearing capacity assessment calculation model;

[0018] Construct a carrying capacity index evaluation system;

[0019] Calculate the carrying capacity of new energy sources to be connected to the distribution network.

[0020] Furthermore, the boundary conditions for setting the bearing capacity assessment calculation model include:

[0021] Analyze the short-circuit current of different new energy sources;

[0022] Calculate the voltage parameters using the voltage drop formula.

[0023] Furthermore, the construction of the bearing capacity index evaluation system includes:

[0024] Calculate the reverse load rate based on the power grid operation mode, and determine the thermal stability constraint, namely the reverse load rate constraint γ, where γ≤80%.

[0025] Determine voltage deviation constraints: Where: V N V is the rated voltage of the power grid, ε is the voltage deviation rate, and V i The voltage at node i;

[0026] Determine voltage fluctuation constraints: d i ≤d max , where d i d represents the percentage of voltage fluctuation at node i under the corresponding operating mode; max The maximum permissible voltage fluctuation value specified in the technical guidelines;

[0027] Determine short-circuit current constraints: I f <I fmax , where I f I is the short-circuit current of the node. fmax This refers to the permissible short-circuit current limit for the power distribution system.

[0028] Furthermore, the calculation of the carrying capacity of new energy access to the distribution network includes:

[0029] Set the objective function for the maximum allowable capacity of new energy sources to be connected to the distribution network: Among them, P newi Let m be the grid-connected power of the i-th renewable energy source, and m be the number of renewable energy sources. Let f be the power factor of the i-th new energy source, and f be the maximum new energy access capacity of the distribution system.

[0030] Determine the constraints of the power flow equations;

[0031] Determine the constraints on new energy output: Among them, P new,i Qnew,i These represent active power output and reactive power output from new energy sources, respectively, P newmax Q newmax These represent the maximum active power output and the maximum reactive power output of new energy sources, respectively. Power factor angle;

[0032] Establish constraints on the assessment indicators for new energy carrying capacity.

[0033] Furthermore, the process of solving the established bearing capacity assessment calculation model using the MSAMRFO algorithm includes:

[0034] Half-uniform initialization: The initial population is divided into two subpopulations, denoted as subpopulation 1 and subpopulation 2. Subpopulation 1 is initialized using the same random initialization method as the traditional MRFO algorithm, while subpopulation 2 will use a half-uniform initialization strategy to ensure the diversity of the entire population.

[0035] Calculate the fitness function: Calculate the fitness value of the population, and add a penalty function value for the population that violates the constraints;

[0036] Updating the location of the population: Based on the foraging behavior of marine manta rays, there are three strategies for updating the location of the population: adaptive chain foraging, adaptive spiral foraging, and adaptive tumbling foraging.

[0037] Population individual updates: If the fitness value of an individual after the iteration is greater than the fitness value before the iteration, the individual after the iteration will replace the individual before the iteration; otherwise, no replacement will be made.

[0038] Convergence determination of the population: The population fitness variance, average clustering distance and clustering threshold are used to determine whether the population has reached global convergence or fallen into a local optimum. If it falls into a local optimum, mutation operation is performed on the local optimum solution with a set mutation probability to make it escape the local optimum solution.

[0039] Furthermore, the mathematical model for the half-uniform initialization strategy is as follows:

[0040]

[0041] Where, p low ,p up These represent the lower and upper bounds of the search space for the population, respectively; rand() represents a random function; p i,1 ,p j,2 These are the random initial values ​​of the i-th individual in subpopulation 1 and the random initial values ​​of the j-th individual in subpopulation 2, respectively. N is the number of individuals in the population.

[0042] Furthermore, the expression for the population fitness variance is:

[0043] In the formula, fit i Let fit be the fitness function value of the i-th individual. avg It is the mean fitness of the population, and fit is the normalization scaling factor;

[0044] The expression for the average aggregation distance of the population is:

[0045] The expression for the aggregation degree threshold is:

[0046] In the formula, Let be the d-th dimension position of the optimal solution found in the t-th iteration of the population search. Let d be the position of the historical optimal solution for the i-th individual; D represents the dimension of the population.

[0047] Furthermore, the mutation probability is between 0.1 and 0.3.

[0048] Furthermore, the formula for performing mutation operations on local optima is as follows:

[0049]

[0050] In the formula, For a local optimal solution, η is a random number that follows a standard normal distribution.

[0051] This invention presents a method for assessing the carrying capacity of renewable energy in power distribution networks based on a multi-strategy improved adaptive manta ray foraging optimization algorithm. The established model considers comprehensive index constraints. Furthermore, this invention employs the multi-strategy improved adaptive manta ray foraging optimization algorithm (MSAMRFO), a more efficient and accurate solution method. In addition, this invention makes several improvements to the general manta ray algorithm (MRFO). First, in terms of population initialization, a semi-uniform initialization strategy is proposed to improve population diversity. Second, a new exponentially varying weighting factor is adopted. MSAMRFO introduces a sigmoid function in chain search and spiral foraging, changing the original random step size to a variable step size, allowing the population to better balance global optimization and local search, while also accelerating the global optimization convergence speed. Finally, to ensure global convergence, it determines whether the population has fallen into a local optimum. The fitness variance and particle aggregation distance of the population are calculated to determine whether the population has fallen into a local optimum, and populations that have fallen into a local optimum are mutated according to a certain mutation probability. Attached Figure Description

[0052] Figure 1This invention provides a solution framework for assessing the renewable energy carrying capacity of distribution networks based on a multi-strategy improved adaptive manta ray foraging optimization algorithm.

[0053] Figure 2 A calculation and evaluation index system for the carrying capacity of new energy sources in power distribution networks;

[0054] Figure 3 A flowchart of the method for assessing the carrying capacity of new energy sources in power distribution networks according to this invention;

[0055] Figure 4 This is a schematic diagram of a 10kV distribution network model. Detailed Implementation

[0056] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0057] The following specific examples illustrate the implementation of this disclosure. Those skilled in the art can easily understand other advantages and effects of this disclosure from the content disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of this disclosure, and not all of them. This disclosure can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this disclosure. It should be noted that, in the absence of conflict, the following embodiments and features in the embodiments can be combined with each other. Based on the embodiments in this disclosure, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this disclosure.

[0058] The present invention provides a method for assessing the renewable energy carrying capacity of distribution networks based on a multi-strategy improved adaptive manta ray foraging optimization algorithm, such as... Figure 1 As shown, it includes the following steps:

[0059] Step S1: Input the distribution network structure data for which the carrying capacity needs to be calculated, and determine the access nodes for new energy sources based on the resource endowment of the distribution network.

[0060] For a distribution network, the geographical location of nodes is limited by natural conditions. Only a few nodes in the distribution network have the ability to plan and build new energy sources. The distribution network operator knows this in advance before planning. Therefore, in order to better reflect the actual situation when conducting carrying capacity assessment calculations, this invention assumes that the nodes for new energy access to the distribution network are known.

[0061] Step S2: Establish a new energy carrying capacity index evaluation system for the distribution network and construct a carrying capacity evaluation calculation model for new energy access.

[0062] Step S2 is carried out in accordance with the requirements of the "Guideline for Assessment of the Power Grid Capacity of Distributed Power Generation" (DL / T2041-2019), and the specific implementation process is as follows:

[0063] Step S201: Set the boundary conditions for the bearing capacity assessment calculation model.

[0064] Furthermore, in a preferred embodiment of this application, the process of setting the boundary conditions of the bearing capacity assessment calculation model includes:

[0065] Step S2011: Analyze the short-circuit current of different new energy types.

[0066] After new energy sources are integrated into the distribution network, the traditional single-power supply mode will be transformed into a dual-power or even multi-power supply mode, and the magnitude, direction, and duration of fault currents in the distribution network will change accordingly. New energy sources can be categorized into three types based on their interface type: inverter interface type (e.g., photovoltaic), synchronous motor type (e.g., gas-fired power), and asynchronous motor type (e.g., wind power). These three types of new energy power sources have different short-circuit current characteristics.

[0067] Inverter-type renewable energy sources lack low-voltage ride-through capability. Once the grid connection point voltage drops below 0.9 pu, the power supply is directly disconnected from the grid, providing zero short-circuit current. When renewable energy sources possess low-voltage ride-through capability, the fault characteristics of the short-circuit current depend on the inverter's control characteristics. To protect the converter, it's crucial to ensure that the short-circuit current during a power system short circuit does not exceed the inverter's limit. Typically, a saturation module is added to the inner current loop of the inverter's control circuit. When a fault occurs, the inner loop reference current is limited. By setting the upper limit of the saturation module, the fault current is limited to an allowable range. Different manufacturers employ different inverter control strategies—primarily manifested in slight variations in the amplitude and duration of the instantaneous short-circuit impact current, and slightly different steady-state values ​​after reaching steady state. However, overall, the maximum short-circuit current after stabilization is approximately 1.2 to 1.5 times the rated current.

[0068] Synchronous generator type new energy: According to relevant research, synchronous generator type new energy power supply can provide up to 10 times the inrush current at the moment of short circuit, and then gradually decays to 2 to 4 times the rated current.

[0069] Induction generator type new energy: According to the research of relevant scholars, the instantaneous short-circuit impact current of asynchronous generator type new energy power supply can be up to 10 times the rated current, and it decays to zero after 10 cycles.

[0070] The conclusion shows that the maximum fault current injection capacity can reach 10 times the rated current. This value can be used for short-circuit calculations to determine the worst-case fault scenario. When analyzing its impact on relay protection, it is necessary to distinguish the magnitude and cycle number of the short-circuit current injected by different types of power supplies. Inverter-type power supplies have smaller short-circuit currents, so the impact is relatively smaller.

[0071] Step S2012: Calculate the voltage parameters according to the voltage drop formula.

[0072] Ignoring the horizontal component of voltage drop, the power flow distribution along the distribution network feeders changes after new energy sources are connected to the grid. According to the voltage drop formula:

[0073]

[0074] In the formula: V0 is the starting voltage, ΔV is the voltage drop, and V is the voltage at the access node. When Pl < P0, R and X are at their maximum values, which is the extreme case at the end of the centralized access line for new energy. If the line voltage does not exceed the limit at this time, other distribution methods can also meet the voltage requirements.

[0075] Step S202: Construct an index system for bearing capacity assessment calculation.

[0076] Furthermore, in a preferred embodiment of this application, constructing the bearing capacity index evaluation system includes:

[0077] Step S2021: Calculate the reverse load rate based on the various operating modes of the power grid, and determine the thermal stability constraint, i.e., the constraint of the reverse load rate.

[0078] Large-scale grid integration of new energy sources drastically alters the original unidirectional power flow, leading to bidirectional power flow and even the transmission of power to the upper-level grid. The transmission and transformation equipment of the distribution network is the subject of thermal stability assessment, and its thermal stability must not exceed prescribed limits. Thermal stability assessment calculates the reverse load rate under various grid operating modes. The formula for calculating the reverse load rate is as follows.

[0079]

[0080] In the formula, P new To contribute to the overall new energy source, P load S represents the total load size. tran This indicates the transformer's capacity.

[0081] The constraints on the reverse load rate are as follows:

[0082] γ≤80% (3)

[0083] Thermal stability assessment should use the maximum value of the reverse load rate γ during the assessment period. maxAs an evaluation indicator, the capacity of new renewable energy sources that can be added to the distribution network to be evaluated is S. m :

[0084] S m =(1-γ) max )*S tran *k r (4)

[0085] In the formula, k r This is the margin factor for the operation of transformer equipment, which is usually 0.8.

[0086] Step S2022: Determine the voltage deviation constraint.

[0087] Ignoring the transverse component of voltage drop, the voltage loss on the line is:

[0088]

[0089] In the formula: ξ is the set of routes, nodes i and j represent the first and last nodes of the route, respectively, and P ij and Q ij These represent the active power and reactive power of the line, R. ij and X ij These represent the resistance and reactance of the power distribution line, respectively; V i Let be the voltage at node i.

[0090] The terminal voltage is the difference between the starting terminal voltage and the line loss voltage, i.e.

[0091]

[0092] The standard GB / T12325-2008, "Power Quality - Supply Voltage Deviation," provides detailed specifications for voltage deviation, with the voltage deviation rate generally set at ε = 0.07. The voltage deviation constraint in the calculation model for the maximum permitted capacity of new energy sources adopts the relevant values ​​specified in the national standard, as detailed in the following formula.

[0093]

[0094] In the formula, V N This is the rated voltage of the power grid.

[0095] Step S2023: Determine voltage fluctuation constraints.

[0096] With the large-scale integration of renewable energy into the distribution network, the randomness of renewable energy output will cause short-term voltage fluctuations in the distribution network. To ensure power quality, voltage fluctuations should not exceed the maximum allowable fluctuation value. The constraints on voltage fluctuations are as follows.

[0097] d i ≤d max (8)

[0098] In the formula, d i d represents the percentage of voltage fluctuation at node i under the corresponding operating mode; max The maximum permissible voltage fluctuation value specified in the technical guidelines is taken as 3% in medium-voltage distribution networks.

[0099] Voltage fluctuations under photovoltaic and wind power integration can be calculated using the following formula.

[0100]

[0101] d PV,k λ represents the voltage fluctuation at node k caused by photovoltaics. P P represents the proportion of the instantaneous power variation of a new energy source caused by natural conditions to its rated output power. PV,j and Q PV,j These represent the active power and reactive power of the photovoltaic output at node j, respectively.

[0102] Let λ be the ratio of the instantaneous power change of a distributed wind turbine caused by factors such as wind speed and direction to its rated power. w Under constant power control mode, the formula for calculating voltage fluctuations caused by new energy wind turbines is summarized as follows:

[0103]

[0104] Among them, P win,j and Q win,j These represent the active power and reactive power of the wind power at node j, respectively.

[0105] Step S2023: Determine the short-circuit current constraint.

[0106] The integration of new energy sources increases short-circuit currents at fault points in the distribution network, posing challenges to the current breaking capacity of circuit breakers and the fault handling capabilities of distribution automation. Short-circuit current verification should be based on the principle that the short-circuit current at each bus node of the system after the integration of new energy sources does not exceed the breaking current limit of the corresponding circuit breaker. The short-circuit current verification formula is as follows.

[0107] I f <I fmax (11)

[0108] I f I is the short-circuit current of the node. fmax This is the limit for the permissible short-circuit current in a power distribution system. In medium-voltage power distribution systems, the short-circuit current limit is generally taken as 20kA.

[0109] In summary, the load-bearing capacity assessment index system is as follows: Figure 2 As shown.

[0110] Step S203: Calculate the carrying capacity of new energy access to the distribution network.

[0111] Furthermore, in a preferred embodiment of this application, calculating the carrying capacity of new energy sources connected to the distribution network includes:

[0112] Step S2031: Set the objective function for the maximum capacity of new energy sources allowed to be connected to the distribution network.

[0113] The maximum capacity of new energy sources allowed to be connected to the distribution network is as follows:

[0114]

[0115] In the above formula, P newi Let m be the grid-connected power of the i-th renewable energy source, and m be the number of renewable energy sources. Let f be the power factor of the i-th new energy source, and f be the maximum new energy access capacity of the distribution system.

[0116] Step S2032: Determine the constraints of the power flow equations.

[0117] Under normal operating conditions, the power flow equations of a distribution network can be expressed as:

[0118]

[0119]

[0120]

[0121] In the above formula, P i Q i Let P be the active and reactive power flowing through node i. new,i+1 Q new,i+1 The active and reactive power outputs of the new energy source at node i+1.

[0122] Step S2033: Determine the power output constraints of new energy sources.

[0123] Due to constraints imposed by natural conditions, the power output of new energy sources is subject to certain limitations.

[0124]

[0125] In the formula, P new,i Q new,i These represent active power output and reactive power output from new energy sources, respectively, P newmax Q newmax These represent the maximum active power output and the maximum reactive power output of new energy sources, respectively. This is the power factor angle.

[0126] Step S2034: Determine the constraints of the new energy carrying capacity assessment indicators.

[0127] The constraints of the carrying capacity assessment index of new energy in the distribution network include the reverse load rate constraint in the thermal stability constraint as shown in Equation (3), the voltage deviation constraint as shown in Equation (8), the voltage fluctuation constraint as shown in Equation (9), and the short-circuit current constraint as shown in Equation (12).

[0128] Step S3: Solve the established bearing capacity assessment calculation model using the MSAMRFO algorithm.

[0129] The MSAMRFO algorithm is used to solve the established bearing capacity assessment calculation model. The specific implementation process is as follows:

[0130] Step S301: Initialize half of the numbers uniformly.

[0131] First, the initial population (with N individuals) is divided into two subpopulations, denoted as subpopulation 1 and subpopulation 2. Subpopulation 1 will use the same random initialization method as the traditional MRFO algorithm, while subpopulation 2 will use a half-uniform initialization strategy to ensure the diversity of the entire population. The mathematical model for the half-uniform initialization strategy is as follows:

[0132]

[0133] Where, p low ,p up These represent the lower and upper bounds of the search space for the population, respectively; rand() represents a random function; p i,1 ,p j,2 These are the random initial values ​​of the i-th individual in subpopulation 1 and the random initial values ​​of the j-th individual in subpopulation 2, respectively.

[0134] This initialization method not only retains the randomness of the traditional MRFO algorithm, but also avoids the concentrated distribution of initial individuals, improves the diversity of the population, and can effectively prevent the population from getting trapped in local optima during the re-iteration process.

[0135] Step S302: Calculate the fitness function.

[0136] The fitness value of the population is calculated based on the objective function (13), and a penalty function value is added to the population that violates the constraint conditions according to the constraint conditions of the carrying capacity calculation model.

[0137] Step S303: Update the location of the population.

[0138] The MSAMRFO algorithm classifies manta rays into three strategies for updating population location based on their foraging behavior: adaptive chain foraging, adaptive spiral foraging, and adaptive tumbling foraging.

[0139] Before each position update, a random number rand is generated. If the random number is greater than 0.5, the adaptive chain foraging method is used to update the position. Otherwise, the adaptive spiral foraging method is used.

[0140] The location update of an adaptive chain-foraging population is shown in the following equation:

[0141]

[0142] In the formula, p d i (t) represents the position of the d-th dimension of the i-th population in the t-th iteration; Let be the position of the optimal solution in the t-th iteration in the d-th dimension; N is the population size; β and α are the weight coefficients, which will be explained below.

[0143] Traditional MRFO algorithms use randomly generated weight coefficients β within a certain range when updating positions, which leads to a decrease in the algorithm's convergence speed. This invention employs a novel exponentially varying weight factor, introducing the Sigmoid function. The weight factor changes with the number of iterations and the convergence rate of the population, allowing the population to better balance global optimization and local search, while also accelerating the global optimization convergence speed.

[0144]

[0145]

[0146] If the generated random number is less than 0.5 and t / T < rand, where t is the current iteration number and T is the total number of iterations, then adaptive spiral foraging strategy 1 will be adopted; if t / T > rand, then adaptive spiral foraging strategy 2 will be adopted.

[0147] The expression for adaptive spiral foraging strategy 1:

[0148]

[0149] The expression for adaptive spiral foraging strategy 2:

[0150]

[0151]

[0152] In the above formula, Let be the randomly generated position in the d-th dimension at iteration t; and are the upper and lower bounds of the variable in the d-th dimension, respectively, and r1 is a random number in the range [0,1].

[0153] During a tumbling hunt, the manta ray uses its current optimal position as a pivot point to tumble to the other side, a mirror image of its current location. The mathematical model is expressed as follows:

[0154]

[0155] In the formula, r2 and r3 are random numbers uniformly distributed on [0,1].

[0156] Step S304, Individual update of the population: If the fitness value of an individual after the iteration is greater than the fitness value before the iteration, the individual after the iteration will replace the individual before the iteration; otherwise, no replacement will be performed.

[0157] After each iteration, some individuals in the population may have new positions that are inferior to their original positions. To accelerate global convergence, it is necessary to determine whether an individual in the population should replace its previous position with its new position, as follows:

[0158]

[0159] Here, `fit()` is the fitness function. The above formula means that if the fitness value of an individual after iteration is greater than the fitness value before iteration, the individual after iteration will replace the individual before iteration; otherwise, no replacement will be made.

[0160] Step S305: Determine the convergence degree of the population.

[0161] To prevent the population from getting trapped in a local optimum, this invention uses population fitness variance, average population aggregation distance, and aggregation degree threshold to determine whether the population has reached global convergence or gotten trapped in a local optimum. If it gets trapped in a local optimum, it is mutated with a certain probability to make it jump out of the local optimum.

[0162] Furthermore, the population fitness variance is shown in the following formula:

[0163]

[0164] In the formula, fit i Let fit be the fitness function value of the i-th individual. avg This is the mean fitness of the population, and `fit` is the normalization scaling factor, which is used to limit δ. 2 The size of fit is shown below.

[0165] fit = max{1, max{|fit i -fit avg |}} (27)

[0166] Furthermore, the average aggregation distance of the population is shown in the following formula:

[0167]

[0168] Furthermore, the population aggregation threshold is shown in the following formula:

[0169]

[0170] In the above formula, Let be the d-th dimension position of the optimal solution found in the t-th iteration of the population search. Let d be the position of the historical optimal solution for the i-th individual; D represents the dimension of the population.

[0171] When the population's fitness variance approaches 0 and the average clustering distance is greater than a clustering threshold, the population is considered to have reached global convergence. Otherwise, when the population's fitness variance approaches 0 and the average clustering distance is less than a clustering threshold, the population is considered to have entered local convergence. A local optimum can be found at a given time. Mutation can be used to help the population escape this local optimum.

[0172] The mutation probability is between 0.1 and 0.3. A mutation operation is performed on the local optimum as follows:

[0173]

[0174] In the formula, For a local optimal solution, η is a random number that follows a standard normal distribution.

[0175] To illustrate the specific implementation process and beneficial effects of the present invention, the method of the present invention is further described below with reference to specific examples:

[0176] Select a 10kV distribution network model, such as Figure 4 As shown, this power distribution network has 45 nodes and 44 lines. The node and line data are shown in Tables 1 and 2. The network has four wind farm connection points: nodes 2, 7, 11, and 15. It also has five photovoltaic connection points: nodes 19, 22, 29, 30, and 40. Using the evaluation method described above, the maximum grid-connected power of each photovoltaic node is calculated to be 3MW, while the maximum grid-connected power of each wind power node is 2MW. This indicates that photovoltaic resources are relatively abundant compared to wind resources in this region.

[0177] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a connection that allows communication between them; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components, unless otherwise explicitly limited. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0178] The above description is merely illustrative of the embodiments of the present invention and is not intended to limit the present invention. For those skilled in the art, any modifications, equivalent substitutions, improvements, etc., made without creative effort within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for assessing the renewable energy carrying capacity of a distribution network based on a manta ray foraging optimization algorithm, characterized in that, Includes the following steps: Input the distribution network structure data for which the carrying capacity needs to be calculated, and determine the access nodes for new energy sources based on the resource endowment of the distribution network; Establish a new energy carrying capacity index evaluation system for distribution networks and construct a carrying capacity evaluation calculation model for new energy access; The established bearing capacity assessment calculation model is solved, including: Half-uniform initialization: The initial population is divided into two subpopulations, denoted as subpopulation 1 and subpopulation 2. Subpopulation 1 is initialized using the same random initialization method as the traditional MRFO algorithm, while subpopulation 2 will use a half-uniform initialization strategy to ensure the diversity of the entire population. Calculate the fitness function: Calculate the fitness value of the population, and add a penalty function value for the population that violates the constraints; The population position is updated using strategies including chain foraging, spiral foraging, and tumbling foraging. Chain foraging and spiral foraging introduce an exponentially changing weight factor of the Sigmoid function to achieve variable step size position updates, while tumbling foraging uses a random step size for position updates. The value of the weight factor is adaptively adjusted by the current iteration number and the distance between the current optimal solution position and the population position. This distance reflects the degree of convergence of the population, so that the weight factor is dynamically determined as the number of iterations and the degree of population convergence change. Population individual updates: If the fitness value of an individual after the iteration is greater than the fitness value before the iteration, the individual after the iteration will replace the individual before the iteration; otherwise, no replacement will be made. Convergence determination of the population: The population fitness variance, average clustering distance and clustering threshold are used to determine whether the population has reached global convergence or fallen into a local optimum. If it falls into a local optimum, mutation operation is performed on the local optimum solution with a set mutation probability to make it escape the local optimum solution.

2. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 1, characterized in that, The aforementioned system for evaluating the carrying capacity of new energy in distribution networks, and the construction of a calculation model for evaluating the carrying capacity of new energy access, include: Define the boundary conditions for the bearing capacity assessment calculation model; Construct a carrying capacity index evaluation system; Calculate the carrying capacity of new energy sources to be connected to the distribution network.

3. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 2, characterized in that, The boundary conditions for the set bearing capacity assessment calculation model include: Analyze the short-circuit current of different new energy sources; Calculate the voltage parameters using the voltage drop formula.

4. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 2, characterized in that, The aforementioned load-bearing capacity index evaluation system includes: Calculate the reverse load rate based on the power grid operation mode, and determine the thermal stability constraint, i.e., the reverse load rate constraint. , ; Determine voltage deviation constraints: ,in: The rated voltage of the power grid. Voltage deviation rate, The voltage at node i; Determine voltage fluctuation constraints: ,in, For nodes under the corresponding operating mode The percentage of voltage fluctuation; The maximum permissible voltage fluctuation value specified in the technical guidelines; Determine short-circuit current constraints: ,in For the short-circuit current of the node, This refers to the permissible short-circuit current limit for the power distribution system.

5. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 2, characterized in that, The calculation of the carrying capacity of new energy access to the distribution network includes: Set the objective function for the maximum allowable capacity of new energy sources to be connected to the distribution network: ,in, Let m be the grid-connected power of the i-th renewable energy source, and m be the number of renewable energy sources. Let be the power factor of the i-th new energy source. This represents the maximum new energy access capacity of the power distribution system. Determine the constraints of the power flow equations; Determine the power output constraints of new energy sources: ,in, These are active power output and reactive power output from new energy sources, respectively. These represent the maximum active power output and the maximum reactive power output of new energy sources, respectively. Power factor angle; Establish constraints on the assessment indicators for new energy carrying capacity.

6. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 1, characterized in that, The mathematical model for the half-uniform initialization strategy is as follows: ； in, These represent the lower and upper bounds of the search space for the population, respectively. Represents a random function; These are the random initial values ​​of the i-th individual in subpopulation 1 and the random initial values ​​of the j-th individual in subpopulation 2, respectively. , N is the number of individuals in the population.

7. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 1, characterized in that, The expression for the population fitness variance is: ; In the formula, Let be the fitness function value of the i-th individual. It is the mean fitness of the population. This is the normalized scaling factor; The expression for the average aggregation distance of the population is: ; The expression for the aggregation degree threshold is: ; In the formula, Let be the d-th dimension position of the optimal solution found in the t-th iteration of the population search. Let d be the position of the historical optimal solution for the i-th individual; D represents the dimension of the population.

8. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 1, characterized in that, The mutation probability is between 0.1 and 0.

3.

9. The method for assessing the carrying capacity of new energy sources in a distribution network based on the manta ray foraging optimization algorithm according to claim 1, characterized in that, The formula for performing mutation operations on local optima is as follows: ； In the formula, This is a locally optimal solution. It is a random number that follows a standard normal distribution.

Images