Frequency strong stability control method for integrated energy distribution network using multi-stage optimization

Abstract

The present application belongs to the technical field of energy distribution network regulation, and particularly relates to a comprehensive energy distribution network frequency strong stability regulation method using multi-stage optimization. With more and more renewable energy generation connected, multiple types of distributed power generation sources are connected to each other to form a multi-region comprehensive energy distribution network, which is increasingly popular. The topology of this type of distribution network is complex, and internal and external disturbances are frequent, so the load frequency is prone to fluctuation and is not easy to control accurately. The present application controls the phase shift angle of the regional output of the comprehensive energy distribution network to control the tie-line power between regions and assist in controlling the frequency of each region to suppress frequency fluctuations. The present application uses a nonlinear multi-stage strong robustness optimization algorithm to perform nonlinear optimization on the power flow controller parameters, and the power flow controller after parameter optimization can alleviate the abnormality caused by load frequency disturbance in the comprehensive energy distribution network and assist in adjusting the frequency deviation and tie-line power deviation to the rated value.

Description

Technical Field

[0001] This invention belongs to the field of energy distribution network control technology, specifically involving a multi-stage optimized method for strong frequency stability control of integrated energy distribution networks. Background Technology

[0002] With the gradual advancement of building a new power system based on renewable energy and the continuous development of integrated energy technologies, more and more renewable energy sources are being integrated into integrated energy distribution networks. Currently, multi-regional integrated energy distribution networks, composed of interconnected distributed generation sources of various types, are becoming increasingly common. However, due to the weak output stability of renewable energy generation, the long and complex circuit topologies of multi-regional distributed access lines, the difficulty in accurate modeling, the strong hysteresis characteristics, and the frequent random interference, maintaining the load frequency stability of integrated energy distribution networks is a very challenging task.

[0003] Frequency drift is mainly caused by sudden imbalances between power generation and power consumption, insufficient power generation reserves, poor synchronization of control and protection devices, or insufficient equipment response. To better regulate load frequency, in addition to traditional secondary frequency controllers, auxiliary frequency control is also widely used in the load frequency regulation of integrated energy distribution networks.

[0004] In current research on load frequency auxiliary regulation, the control method and parameter selection of the load frequency auxiliary regulation controller play a decisive role in the effectiveness of auxiliary regulation. Different controller methods show significant differences in load frequency auxiliary regulation. Currently, the main approach is to control the power of the power interconnection lines between control areas, adjust the power in the power interconnection lines, and alleviate high-frequency anomalies to achieve the auxiliary regulation effect. Due to the increasing complexity of integrated energy distribution networks, the load frequency auxiliary regulation controller needs to have strong stability and a short response time to achieve the expected auxiliary regulation effect. Therefore, for complex integrated energy distribution networks, a highly stable and rapidly responsive load frequency auxiliary regulation controller is needed to obtain ideal dynamic performance.

[0005] For complex integrated energy distribution networks, obtaining optimal controller parameters requires high-performance heuristic algorithms to optimize the load frequency auxiliary control controller parameters. However, general heuristic algorithms are more prone to revealing their shortcomings when dealing with complex systems, such as weak search capabilities, slow iteration speeds, and a tendency to get trapped in local optima. These shortcomings prevent the optimized load frequency auxiliary control controller from achieving the expected control effect. Therefore, to ensure the load frequency auxiliary control controller achieves rational performance, a highly robust optimization algorithm needs to be constructed to optimize the controller parameters and achieve the optimal control effect. Summary of the Invention

[0006] This invention proposes a multi-stage optimized method for strong frequency stability control of integrated energy distribution networks, targeting load frequency regulation in integrated energy distribution networks.

[0007] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0008] A multi-stage optimized method for strong frequency stability control of integrated energy distribution networks includes the following steps:

[0009] S1. Obtain the power of the regional tie line of the integrated energy distribution network and construct a power flow controller adapted to the auxiliary control of load frequency regulation of the integrated energy distribution network;

[0010] S2. Obtain the frequency deviation and tie-line power deviation of each region, construct the time-squared error integral performance evaluation index, and use this index as the objective function for minimizing the error.

[0011] S3. Construct a nonlinear multi-stage robust optimization algorithm with nonlinear weights and adaptive parameters for a power flow controller for load frequency regulation auxiliary control of integrated energy distribution network.

[0012] S4. A nonlinear multi-stage robust optimization algorithm is applied to perform nonlinear multi-stage iterative optimization of the power flow controller parameters within the constraint range; through multiple optimization stages, the parameters are optimized quickly and accurately; the optimized parameters are then assigned to the power flow controller for load frequency regulation control of the integrated energy distribution network to perform auxiliary control of load frequency regulation and achieve strong frequency stability.

[0013] As a further preferred embodiment of the present invention, the integrated energy distribution network comprises n regions, each region comprising a power source and a secondary frequency regulation controller, and a power flow controller between every two regions. The control of the power flow controller in S1 specifically includes the following steps:

[0014] S1.1 Obtain the voltage and voltage phase shift angle of region i and region j, and calculate the total power P shared from region i to region j. s :

[0015]

[0016] In the formula, U i U j The voltages in regions i and j are δ, respectively. i δ j The voltage phase shift angles for regions i and j are x and x, respectively. tie is the tie line impedance, and i and j are the numbers of any two different regions in the integrated energy distribution network;

[0017] S1.2 The total power shared between region i and region j is divided into active power and reactive power. Calculate the active power P shared between region i and region j. αij :

[0018]

[0019] S1.3 Calculate the active power P shared between region i and region j under the auxiliary control of the power flow controller. βij :

[0020]

[0021] In the formula, The phase shift angle is obtained through the power flow controller;

[0022] S1.4 Calculate the tie-line power deviation P under power flow controller-assisted control. vij :

[0023]

[0024] In the formula, T ij For the power system gain, Δδ i , Δδ j The voltage phase shift angle deviations Δf and j are respectively for regions i and j. i , Δf j Let be the frequency deviations for regions i and j, respectively;

[0025] S1.5 Calculate the change in tie-line power deviation P between regions i and j in the frequency domain. tieij :

[0026]

[0027] In the formula, ΔF i (s), ΔF j (s) represents the frequency deviation between region i and region j in the frequency domain;

[0028] S1.6. The phase shift angle is controlled by a power flow controller to change the tie-line power deviation; the input of the power flow controller is the frequency deviation of region i, and the output is the change in phase shift angle; calculate the phase shift angle generated by the power flow controller in region i.

[0029]

[0030] In the formula, For the power flow controller gain, T S The power flow controller time constant and the power flow controller gain are given. It has range constraints, and the gain of the power flow controller is adjusted according to the constraint range;

[0031]

[0032] In the formula, These represent the maximum and minimum gain values ​​for the power flow controller.

[0033] Furthermore, as a preferred embodiment of the present invention, each regional power tie line of the integrated energy distribution network includes the power flow controller, and the time-squared error integral performance evaluation index is used as the error minimization optimization objective. The objective function is established through formula (8):

[0034]

[0035] In the formula, J is the time-squared error integral performance evaluation index value, t0 is the start time of the control cycle, t1 is the end time of the control cycle, t is the control cycle time, and n is the number of integrated energy distribution network areas.

[0036] Furthermore, as a preferred embodiment of the present invention, step S3 specifically includes the following steps:

[0037] S3.1 Initialize the system frequency target matrix and the system frequency individual matrix; for each system frequency element X in the system frequency individual matrix y,z Perform initialization;

[0038]

[0039] In the formula, y represents the population to which the system frequency element belongs, z represents the dimension of the population to which the system frequency element belongs, and X represents the dimension of the population to which the system frequency element belongs. min X max Let be the upper and lower bounds of the matrix elements. It is a random vector belonging to [0,1].

[0040] Construct the system frequency individual matrix Pr:

[0041]

[0042] In the formula, q is the total number of populations and d is the total number of dimensions;

[0043] Initialize the system frequency target matrix, calculate the fitness of each element in the system frequency individual matrix, and construct the system frequency target matrix E1 by replicating the individual with the best fitness q times:

[0044]

[0045] In the formula, For the system frequency target matrix elements;

[0046] S3.2. Construct nonlinear control parameters, which are used to balance different optimization stages, and obtain the nonlinear control parameter w according to the number of iterations K:

[0047]

[0048] where K is the number of iterations, and M is the maximum number of iterations;

[0049] S3.3. First-stage optimization; when K<1 / 3M, the first-stage optimization is entered, which is the initial stage of parameter optimization;

[0050]

[0051] where, is the step size, is the individual position, is the vector form of the system frequency target matrix, denotes entry-wise multiplication operation, is a vector composed of random numbers generated by random walk, is a vector composed of random numbers uniformly distributed between 0 and 1;

[0052] S3.4. Second-stage optimization; when 1 / 3M<K<2 / 3M, this stage occurs in the middle of the optimization process and is the in-depth stage of parameter optimization; in this stage, the population is divided into two parts for optimization; the first half of the population is optimized;

[0053]

[0054] where, is a vector composed of random numbers of stable distribution;

[0055] the second half of the population is optimized;

[0056]

[0057] where C is the adaptive parameter of the step size;

[0058] S3.5. Third-stage optimization; when K>2 / 3M, the third-stage optimization is entered, which is the later stage of parameter optimization;

[0059]

[0060] S3.6. Resolve the error disturbances that will occur during the operation of the algorithm to achieve better optimization accuracy;

[0061]

[0062] where r is a random number in [0,1], Let r1 and r2 be binary vectors, where r1 and r2 are the random exponents of an individual.

[0063] S3.7 Update the system frequency target matrix. For each individual in the target matrix, calculate its fitness. If the fitness is better than the fitness of the corresponding position in the original target matrix, then replace the corresponding individual in the original target matrix with that individual. Then calculate the fitness of the best individual in the entire system frequency target matrix. If the requirement is met, the algorithm ends; otherwise, continue iterating.

[0064] Furthermore, as a preferred embodiment of the present invention, step S4 specifically includes the following steps:

[0065] S4.1 Based on the objective function and parameter constraint range in S2, the nonlinear multi-stage robust optimization algorithm in S3 is applied to perform nonlinear iterative optimization of the load frequency-assisted power flow controller parameters in S1.6 within the constraint range given in S2.

[0066] S4.2. The parameters optimized by the nonlinear multi-stage robust optimization algorithm in S3 are assigned to the power flow controller for load frequency auxiliary regulation of the integrated energy distribution network.

[0067] S4.3 When a load disturbance occurs in the integrated energy distribution network, the power flow controller for load frequency auxiliary regulation established in S1.6 can achieve a rapid response, perform auxiliary regulation on the load frequency, restore the system frequency deviation and tie-line power deviation to the rated value, and maintain the frequency deviation and tie-line power deviation of the entire integrated energy distribution network within the specified range.

[0068] The multi-stage optimized integrated energy distribution network frequency strong stability control method described in this invention has the following technical advantages compared with existing technologies:

[0069] (1) The power flow controller of the present invention assists in the regulation of the load frequency of the integrated energy distribution network. When the integrated energy distribution network experiences load disturbance, it responds quickly. By controlling the regional output phase shift angle of the integrated energy distribution network, it controls the power of the inter-regional tie line, assists in controlling the frequency of each region, suppresses frequency fluctuations, and alleviates the abnormality caused by load frequency disturbance. It has strong adaptive capability.

[0070] (2) This invention constructs a nonlinear multi-stage robust optimization algorithm that includes nonlinear weights and adaptive control parameters. Since the nonlinear weights w can balance the global search stage and the local development stage of the algorithm, and the adaptive control parameters C can adjust the step size of the individual, the algorithm has strong global search capability and local development capability, and has efficient parameter optimization capability.

[0071] (3) This invention utilizes a nonlinear multi-stage robust optimization algorithm to perform nonlinear multi-stage optimization of the parameters of the designed load frequency-assisted control power flow controller. It can optimize the power flow controller parameters in a short time. The optimized power flow controller can achieve fast response and assist in adjusting the system frequency deviation and tie line power deviation to the rated value. It has strong stability and strong adaptive capability. Attached Figure Description

[0072] Figure 1 This is a schematic diagram of a two-region integrated energy distribution network system in one embodiment of the present invention;

[0073] Figure 2 This is a schematic diagram of a power flow controller in one embodiment of the present invention;

[0074] Figure 3 This is a schematic diagram of the nonlinear multi-stage robust optimization algorithm in one embodiment of the present invention. Detailed Implementation

[0075] The present invention will now be described in further detail with reference to the accompanying drawings.

[0076] A multi-stage optimized integrated energy distribution network frequency strong stability control method is adopted, including the following steps: S1, Appendix Figure 1 In this embodiment, a two-region integrated energy distribution network is described. The system comprises two regions, n=2, i=1, j=2. Each region has one photovoltaic power generation system and one non-reheat thermal power generation system. A PID controller is used as the secondary frequency regulation controller for each region. When a load disturbance occurs in the integrated energy distribution network, each region is affected. The two regions are connected via a power tie line to achieve inter-regional power transmission. Simultaneously, a power flow controller is added to the power tie line between the two regions to assist in the regulation of the load frequency. The power flow controller controls the power of the inter-regional tie line by controlling the phase shift angle of the regional output of the integrated energy distribution network, thereby assisting in the frequency control of each region. (Appendix) Figure 1 In the diagram, R1 is the droop coefficient of the non-reheat thermal power generation system, B1 and B2 are the frequency deviation coefficients of region 1 and region 2, and T... 12 ΔF1 and ΔF2 are the power line synchronization coefficients, ΔF1 and ΔF2 are the frequency deviations between region 1 and region 2, and ΔP is the power line synchronization coefficient. tie12 α represents the power deviation of the tie line between Region 1 and Region 2. 12 This is the power conversion factor.

[0077] S1.1 Obtain the voltage and voltage phase shift angle of region i and region j, and calculate the total power P shared from region i to region j. s .

[0078]

[0079] In the formula, U i U j The voltages in regions i and j are δ, respectively. i δ j The voltage phase shift angles for regions i and j are x and x, respectively. tie is the tie line impedance, and i and j are the numbers of any two different areas in the integrated energy distribution network.

[0080] S1.2 The total power shared between region i and region j is divided into active power and reactive power. Calculate the active power P shared between region i and region j. αij .

[0081]

[0082] S1.3 Calculate the active power P shared between region i and region j under the auxiliary control of the power flow controller. βij .

[0083]

[0084] In the formula, This is the phase shift angle obtained through the power flow controller.

[0085] S1.4 Calculate the tie-line power deviation P under power flow controller-assisted control. vij .

[0086]

[0087] In the formula, T ij For the power system gain, Δδ i , Δδ j The voltage phase shift angle deviations Δf and j are respectively for regions i and j. i , Δf j Let be the frequency deviations for regions i and j, respectively.

[0088] S1.5 Calculate the change in tie-line power deviation P between regions i and j in the frequency domain. tieij .

[0089]

[0090] In the formula, ΔF i (s), ΔF j (s) represents the frequency deviation between region i and region j in the frequency domain.

[0091] S1.6. The phase shift angle is controlled by a power flow controller, thereby changing the tie-line power deviation. The input of the power flow controller is the frequency deviation in region i, and the output is the change in phase shift angle. Calculate the phase shift angle generated by the power flow controller in region i.

[0092]

[0093] In the formula, For the power flow controller gain, T S The time constant of the power flow controller. The gain of the power flow controller. It has range constraints, and the gain of the power flow controller is adjusted according to the constraint range.

[0094]

[0095] In the formula, These represent the maximum and minimum gain values ​​for the power flow controller.

[0096] S2. In the embodiment, the inter-regional power tie line of the integrated energy distribution network includes the power flow controller. The time-squared error integral performance evaluation index is used as the error minimization optimization target, and the objective function is established by formula (8).

[0097]

[0098] In the formula, J is the time-squared error integral performance evaluation index value, t0 is the start time of the control cycle, t1 is the end time of the control cycle, t is the control cycle time, and n is the number of integrated energy distribution network areas.

[0099] S3, Appendix Figure 3 The flowchart illustrates a nonlinear multi-stage robust optimization algorithm. For load frequency-assisted regulation in integrated energy distribution networks, a nonlinear multi-stage robust optimization algorithm incorporating nonlinear weighted adaptive control parameters is constructed. This robust optimization algorithm is applied to perform nonlinear multi-stage optimization of the load frequency-assisted regulation power flow controller parameters within constraints. The steps are as follows:

[0100] S3.1 Initialize the system frequency target matrix and the system frequency individual matrix.

[0101] For each system frequency element X in the system frequency individual matrix y,z Perform initialization.

[0102]

[0103] In the formula, y represents the population to which the system frequency element belongs, z represents the dimension of the population to which the system frequency element belongs, and X represents the dimension of the population to which the system frequency element belongs. min X maxare the upper and lower limits of the matrix elements, is a random vector belonging to [0,1].

[0104] Construct the system frequency individual matrix Pr.

[0105]

[0106] In the formula, q is the total population size, d is the total number of dimensions, and in the embodiment, q=35 and d=4.

[0107] Initialize the system frequency target matrix, calculate the fitness of each element in the system frequency individual matrix, and copy q copies of the individual with the optimal fitness to form the system frequency target matrix El.

[0108]

[0109] In the formula, are the elements of the system frequency target matrix.

[0110] S3.2. Construct nonlinear control parameters. The nonlinear parameters are used to balance different optimization stages, and the nonlinear control parameter w is obtained according to the number of iterations K.

[0111]

[0112] In the formula, K is the number of iterations, and M is the maximum number of iterations.

[0113] S3.3. First-stage optimization. When K<1 / 3M, the process enters the first-stage optimization, which is the initial stage of parameter optimization.

[0114]

[0115] In the formula, is the step size, is the individual position, the vector form of the system frequency target matrix is is the entry-wise multiplication operation, is a vector composed of random numbers generated by random walk. is a vector composed of random numbers uniformly distributed between [0,1].

[0116] S3.4. Second-stage optimization. When 1 / 3M<K<2 / 3M, this stage occurs in the middle of the optimization process and is the in-depth stage of parameter optimization. In this stage, the population is divided into two parts for optimization. The first half of the population is optimized.

[0117]

[0118] In the formula, It is a vector composed of stably distributed random numbers.

[0119] The latter half of the population was optimized.

[0120]

[0121] In the formula, C is the adaptive parameter of the step size.

[0122] S3.5 Three-stage optimization. When K > 2 / 3M, the three-stage optimization begins, which is the later stage of parameter optimization.

[0123]

[0124] S3.6. Address the error disturbances that may occur during the algorithm's operation to achieve better optimization accuracy.

[0125]

[0126] In the formula, r is a random number in [0,1]. Let r1 and r2 be binary vectors, where r1 and r2 are the random exponents of the individual.

[0127] S3.7 System frequency target matrix update: For each individual in the system frequency target matrix, calculate its fitness. If the fitness is better than the fitness of the corresponding position in the original target matrix, then replace the corresponding individual in the original target matrix with that individual. Then calculate the fitness of the best individual in the entire system frequency target matrix. If it meets the requirements, the algorithm ends; otherwise, continue iterating.

[0128] S4, in this embodiment, S4 specifically includes the following steps:

[0129] S4.1 Based on the objective function and parameter constraint range in S2, the nonlinear multi-stage robust optimization algorithm in S3 is applied to perform nonlinear iterative optimization of the load frequency-assisted power flow controller parameters in S1.6 within the constraint range given in S2.

[0130] S4.2. The parameters optimized by the nonlinear multi-stage robust optimization algorithm in S3 are assigned to the power flow controller for load frequency auxiliary regulation of the integrated energy distribution network.

[0131] S4.3 When a load disturbance occurs in the integrated energy distribution network, the power flow controller for load frequency auxiliary regulation established in S1.6 can achieve a rapid response, perform auxiliary regulation on the load frequency, restore the system frequency deviation and tie-line power deviation to the rated value, and maintain the frequency deviation and tie-line power deviation of the entire integrated energy distribution network within the specified range.

[0132] The power flow controller of this invention provides auxiliary regulation of the load frequency of an integrated energy distribution network. It responds rapidly to load disturbances in the integrated energy distribution network, controlling the regional output phase shift angle to regulate the power of inter-regional tie lines, thus assisting in frequency control of each region, suppressing frequency fluctuations, and mitigating anomalies caused by load frequency disturbances. It exhibits strong adaptive capability. This invention constructs a nonlinear multi-stage robust optimization algorithm incorporating nonlinear weights and adaptive control parameters. Since the nonlinear weights w balance the global search and local development stages of the algorithm, and the adaptive control parameters C adjust the step size of individual parameters, the algorithm possesses strong global search and local development capabilities, resulting in efficient parameter optimization. This invention utilizes a nonlinear multi-stage robust optimization algorithm to perform nonlinear multi-stage optimization of the designed load frequency auxiliary regulation power flow controller parameters. This enables optimization of the power flow controller parameters in a short time. The optimized power flow controller achieves rapid response, assisting in adjusting system frequency deviation and tie line power deviation to rated values, exhibiting strong stability and strong adaptive capability.

[0133] The specific implementation schemes described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific implementation schemes of the present invention and are not intended to limit the scope of the present invention. Any equivalent changes and modifications made by those skilled in the art without departing from the concept and principles of the present invention should fall within the scope of protection of the present invention.

Claims

1. A multi-stage optimized integrated energy distribution network frequency strong stability control method, characterized in that, comprises the following steps: S1. Acquire the tie line power of the integrated energy distribution network region, and construct a power flow controller adapted to the auxiliary control of load frequency regulation of the integrated energy distribution network; S2. Acquire the frequency deviation of each region and the tie line power deviation, construct a time multiplied squared error integral performance evaluation index, and take this index as the error minimization objective function; S3. Construct a nonlinear multi-stage strong robustness optimization algorithm containing nonlinear weights and adaptive parameters for the power flow controller for the auxiliary control of load frequency regulation of the integrated energy distribution network; S4. Apply the nonlinear multi-stage strong robustness optimization algorithm to perform nonlinear multi-stage iterative optimization on the parameters of the power flow controller within the constraint range; optimize the parameters quickly and accurately through multiple optimization stages; assign the optimized parameters to the power flow controller for load frequency regulation control of the integrated energy distribution network to perform auxiliary control of load frequency regulation, so as to achieve strong frequency stability; The integrated energy distribution network comprises n regions, each region comprises a power generation source and a secondary frequency modulation controller, a power flow controller is arranged between every two regions, and the control of the power flow controller in S1 specifically comprises the following steps: S1.1 Obtain the voltage and voltage phase shift angle of region i and region j, and calculate the total power P shared from region i to region j. s : （1） In the formula, U i U j The voltages in regions i and j are δ, respectively. i δ j The voltage phase shift angles for regions i and j are x and x, respectively. tie is the tie line impedance, and i and j are the numbers of any two different regions in the integrated energy distribution network; S1.2 The total power shared between region i and region j is divided into active power and reactive power. Calculate the active power P shared between region i and region j. αij : （2） S1.3 Calculate the active power P shared between region i and region j under the auxiliary control of the power flow controller. βij : （3） wherein φ in the formula is the phase shift angle obtained by the power flow controller; S1.4 Calculate the tie-line power deviation P under power flow controller-assisted control. vij : （4） In the formula, T ij For the power system gain, Δδ i , Δδ j The voltage phase shift angle deviations Δf and Δf represent the voltage phase shift angle deviations in regions i and j, respectively. i , Δf j Let be the frequency deviations for regions i and j, respectively; S1.5 Calculate the change in tie-line power deviation P between regions i and j in the frequency domain. tieij : （5） In the formula, ΔF i (s), ΔF j (s) represents the frequency deviation between region i and region j in the frequency domain; S1.

6. Control the change of the phase shift angle through the power flow controller, so as to change the tie line power deviation; the input of the power flow controller is the frequency deviation of region i, and the output is the change of the phase shift angle; calculate the phase shift angle generated by the power flow controller in region i; （6） In the formula, K φ For the power flow controller gain, T S Let K be the time constant of the power flow controller, and K be the gain of the power flow controller. φ It has range constraints, and the gain of the power flow controller is adjusted according to the constraint range; （7） wherein Kmax φ and Kmin φ in the formula are the maximum value and the minimum value of the gain of the power flow controller respectively.

2. The method for strong frequency stability control of integrated energy distribution networks using multi-stage optimization as described in claim 1, characterized in that, Each regional power tie line of the integrated energy distribution network is provided with the power flow controller, the time multiplied squared error integral performance evaluation index is taken as the error minimization optimization target, and the objective function is established through formula (8): （8） wherein J in the formula is the value of the time multiplied squared error integral performance evaluation index, t0 is the start time of the control period, t1 is the end time of the control period, t is the time of the control period, and n is the number of regions of the integrated energy distribution network.

3. The integrated energy distribution network frequency strong stability control method using multi-stage optimization as described in claim 2, characterized in that, S3 specifically comprises the following steps: S3.1 Initialize the system frequency target matrix and the system frequency individual matrix; for each system frequency element X in the system frequency individual matrix y,z Perform initialization; （9） In the formula, y represents the population to which the system frequency element belongs, z represents the dimension of the population to which the system frequency element belongs, and X represents the dimension of the population to which the system frequency element belongs. min X max Let be the upper and lower bounds of the matrix elements. It is a random vector belonging to [0,1]. Construct a system frequency individual matrix Pr: （10） wherein q in the formula is the total number of populations, and d is the total number of dimensions; Initialize a system frequency target matrix, calculate the fitness of each element in the system frequency individual matrix, and replicate the individual with the optimal fitness for q times to form a system frequency target matrix El; （11） In the formula, For the system frequency target matrix elements; S3.

2. Construct a nonlinear control parameter, the nonlinear parameter is used to balance different optimization stages, and the nonlinear control parameter w is obtained according to the number of iterations K: （12） wherein K in the formula is the number of iterations, and M is the maximum number of iterations; S3.

3. First-stage optimization; when K < 1 / 3M, the process enters the first-stage optimization, which is the initial stage of parameter optimization; (13) In the formula, Step size, For individual location, The vector form of the system frequency target matrix is: This is an item-by-item multiplication operation. It is a vector composed of random numbers generated by a random walk. It is a vector of random numbers uniformly distributed between [0,1]. S3.

4. Second-stage optimization; when 1 / 3M < K < 2 / 3M, this stage occurs in the middle of the optimization process, which is the in-depth stage of parameter optimization; in this stage, the population is divided into two parts for optimization; the first half of the population is optimized; （14） In the formula, It is a vector composed of stably distributed random numbers; the second half of the population is optimized; （15） wherein C in the formula is the adaptive parameter of the step size; S3.5 Three-stage optimization; When K>2 / 3M, enter the three-stage optimization, which is the later stage of parameter optimization; （16） S3.

6. Address the error disturbances that may occur during the algorithm's operation to achieve better optimization accuracy; （17） In the formula, r is a random number in [0,1]. Let r1 and r2 be binary vectors, where r1 and r2 are the random exponents of an individual. S3.7 Update the system frequency target matrix. For each individual in the target matrix, calculate its fitness. If the fitness is better than the fitness of the corresponding position in the original target matrix, then replace the corresponding individual in the original target matrix with that individual. Then calculate the fitness of the best individual in the entire system frequency target matrix. If the requirement is met, the algorithm ends; otherwise, continue iterating.

4. The integrated energy distribution network frequency strong stability control method using multi-stage optimization as described in claim 3, characterized in that, S4 specifically includes the following steps: S4.1 Based on the objective function and parameter constraint range in S2, the nonlinear multi-stage robust optimization algorithm in S3 is applied to perform nonlinear iterative optimization of the load frequency-assisted power flow controller parameters in S1.6 within the constraint range given in S2. S4.

2. The parameters optimized by the nonlinear multi-stage robust optimization algorithm in S3 are assigned to the power flow controller for load frequency auxiliary regulation of the integrated energy distribution network. S4.3 When a load disturbance occurs in the integrated energy distribution network, the power flow controller for load frequency auxiliary regulation established in S1.6 can achieve a rapid response, perform auxiliary regulation on the load frequency, restore the system frequency deviation and tie-line power deviation to the rated value, and maintain the frequency deviation and tie-line power deviation of the entire integrated energy distribution network within the specified range.

Images