A Microgrid Coordination Control Method Based on Predictive Control of Nonlinear System Behavior Trajectory

Abstract

The present invention relates to the technical field of microgrids, and particularly to a coordinated control method for a microgrid based on the prediction control of the behavior trajectory of a nonlinear system. The control method includes the following steps: exciting the dynamic response of the microgrid units by injecting an excitation signal, collecting historical operation data and real-time operation data, and respectively constructing a historical behavior description and a real-time behavior description that characterize the nonlinear behavior of the units; based on the historical behavior description and the real-time behavior description, generating an optimal control law by solving an optimization problem that includes the system stability objective, and superimposing the control instruction onto the unit control loop; according to the dynamic harmonic characteristics of the unit connection point, screening the units that need to update the behavior description, and periodically updating their historical behavior data to adapt to the dynamic changes of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of microgrids, and particularly to a coordinated control method for microgrids based on the prediction control of the behavior trajectory of a nonlinear system. Background Art

[0002] Microgrids can fully promote the large-scale access of renewable energy, achieve a highly reliable supply to local loads, and enable the transition of traditional power grids to smart grids, and have achieved remarkable development. Due to the dynamic interaction between elements such as new energy and weak power grids, stability problems such as high-frequency oscillations are likely to occur, seriously affecting the safe and stable operation of microgrids and severely restricting the consumption capacity of new energy.

[0003] At present, the coordinated control methods for microgrids can be divided into three categories: optimizing control parameters, additional control methods, and replacing the original control.

[0004] (1) The principle of the method of optimizing control parameters is that the stability problem of the microgrid is closely related to the control parameters. By optimizing the control parameters of different control objects and different control levels, the effect of stabilizing the system operation can be achieved;

[0005] (2) The additional control method is to introduce additional control on the basis of the original control method and control structure of the equipment, and adjust the control characteristics of the equipment through the additional control, so as to achieve the stable operation of the system;

[0006] (3) Replacing the original control means replacing the original control strategy with a new control strategy, so that the equipment exhibits new control characteristics, and further eliminating the stability problems brought about under the original control.

[0007] Among the existing microgrid coordinated control methods, the methods of optimizing control parameters and additional control methods require detailed modeling of the units, and do not consider the differences between the actual units and the theoretical aggregation models. When the system parameters change greatly, the stable operation of the system cannot be achieved, and the robustness is weak; replacing the original control requires a complete replacement of the controller, which is difficult in practical applications, and there is a possibility of new stability problems after replacement. Summary of the Invention

[0008] In view of the above-mentioned prior art, the present invention aims to provide a coordinated control method for microgrids based on the prediction control of the behavior trajectory of a nonlinear system, mainly solving the technical problems existing in the above-mentioned background art.

[0009] To achieve the above object, the technical solution of the embodiment of the present invention is realized as follows:

[0010] A coordinated control method for microgrids based on the prediction control of the behavior trajectory of a nonlinear system, the control method comprising the following steps:

[0011] The dynamic response of the microgrid unit is excited by injecting an excitation signal, and historical operation data and real-time operation data are collected to construct a historical behavior description and a real-time behavior description that characterize the non-linear behavior of the unit respectively;

[0012] Based on the historical behavior description and the real-time behavior description, an optimal control law is generated by solving an optimization problem that includes the system stability objective, and the control instruction is superimposed on the unit control loop;

[0013] According to the dynamic harmonic characteristics of the unit connection point, the units that need to update the behavior description are screened, and their historical behavior data is periodically updated to adapt to the dynamic changes of the system.

[0014] Optionally, the dynamic response of the microgrid unit is excited by injecting an excitation signal, specifically including: at moment, a white noise signal with a per-unit noise power of is injected into the input signal of the unit non-linear system to trigger the persistent excitation of the non-linear system output . Based on the sampling time for , are sampled to obtain a non-linear historical measurement sequence with a length of ,

[0015] where the non-linear historical measurement sequence includes an input sequence and an output sequence ;

[0016] where represents the input data at time i, and the input data simultaneously includes the d-axis current additional reference value signal of the converter and the q-axis current additional reference value signal , represents an input sequence composed of data with a length of T starting from time i, represents the output data at time i, represents an output sequence composed of data with a length of T starting from time i.

[0017] Optionally, before establishing a non-linear behavior trajectory prediction model based on the historical behavior description and the real-time behavior description, it also includes constructing a non-linear system of the new energy unit, and describing the constructed non-linear system of the new energy unit as:

[0018]

[0019] After linearizing the non-linear system of the new energy unit, it is described as:

[0020]

[0021] Among them, and are the vector fields of the new energy unit, represents the time instance of the continuous-time function of the new energy unit, represents the state sequence composed of data with length T, represents the input sequence composed of data with length T, represents the increment of the state sequence at time k, represents the increment of the input sequence at time k, represents the increment of the state sequence at time k + 1, represents the increment of the input sequence at time k + 1, represents the derivative of x with respect to f under the sequences and ; represents the derivative of x with respect to h under the sequences and ; represents the derivative of u with respect to f under the sequences and ; represents the derivative of u with respect to h under the sequences and , where represents the state quantity of the nonlinear system at time k, represents the output quantity of the nonlinear system at time k, represents the increment of the output sequence at time k + 1.

[0022] Optionally, obtain the descriptions of the historical behavior and real-time behavior of the nonlinear system, specifically including:

[0023] Based on the input sequence and the output sequence, construct an input Hankel matrix and an output Hankel matrix respectively, both of which have L rows and T - L + 1 columns;

[0024] Horizontally divide both the input Hankel matrix and the output Hankel matrix into two parts. The first rows are used to implicitly estimate the initial state of the system, and the last N rows are used to predict the future trajectory of the system. The division results are:

[0025]

[0026] Among them, , , , are the sub-matrices after the corresponding Hankel matrix division. The superscript represents the initial recording time of the data sequence, represents the input Hankel matrix, Denote the output Hankel matrix;

[0027] The historical behavior data matrix at a moment is: , and use the historical behavior data matrix as the historical behavior of the nonlinear system;

[0028] Record the unit operation data starting from the k-th moment, and the time length of the recorded data is , the additional reference value signal of the d-axis current and the additional reference value signal of the q-axis current , the active power at the grid connection point of the unit and the reactive power , and construct the real-time behavior data matrix at the k-th moment as: , where and are the input sequences starting from the k-th moment and with a length of respectively, and is the data sequence with a length of N to be predicted, and use the real-time behavior data matrix as the real-time behavior description.

[0029] Optionally, the optimization problem is expressed as:

[0030]

[0031]

[0032] where, represents the input of the nonlinear system obtained by solving the optimization problem at the k-th moment, represents the output of the nonlinear system obtained by solving the optimization problem at the k-th moment, represents the vector composed of from the i-th moment to the i + j - 1-th moment, represents the vector composed of from the i-th moment to the i + j - 1-th moment, is the set value of the input of the nonlinear system at the k-th moment, is the set value of the output of the nonlinear system at the k-th moment, o represents a row vector, represents a column vector with a length of n + 1, is the control cost matrix, and are both output cost matrices, represents the output response value that the system should finally reach, is the auxiliary slack variable, and are regularization parameters, and g is the decision variable of the above optimization problem, denotes the Kronecker product.

[0033] Optionally, use the MATLAB solver or implement the solution of the optimization problem through programming to obtain the optimal control law , and inject the reference values of the dq-axis currents of the converters and to achieve microgrid oscillation suppression.

[0034] Optionally, according to the dynamic harmonic characteristics of the grid connection points of the units, screen the units that need to update the behavior description, specifically including: calculating the proportion of the harmonic content at the grid connection points of each new energy unit, and selecting the units for the timing update algorithm according to the cumulative sum threshold truncation method.

[0035] Optionally, after calculating the proportion of the harmonic content of the i-th new energy unit, sort the proportions of the harmonic content of all new energy units from largest to smallest to obtain a sorted list, perform cumulative calculation on the sorted proportions of the harmonic content, and when the cumulative calculation result ≥ 50%, select the units included in the cumulative calculation process as the set of units that need to update the algorithm.

[0036] Optionally, after Tupdate time, recalculate the optimal control law and the set of units that need to update the algorithm.

[0037] The beneficial effects of the present invention are as follows: Based on the microgrid behavior trajectory data, a non-linear system behavior trajectory predictor is constructed, and an additional controller is constructed based on the control objective to achieve microgrid coordinated control and stable operation. The non-linear system behavior trajectory predictor in the invention is constructed based on the microgrid operation data, does not require accurate system modeling, avoids the dependence on detailed modeling of the units in the traditional method, can adapt to the dynamic changes of system parameters, and has strong robustness. The microgrid controller in the invention only superimposes a certain perturbation on the original controller, and the controller output will be set to zero after the system operates stably. Therefore, it is not necessary to completely replace the original controller, which reduces the implementation complexity and cost, and at the same time avoids the new stability problems that may be brought by replacing the controller. Through the cumulative sum threshold truncation method of the harmonic content proportion, the units that have the greatest impact on the system stability can be efficiently screened out, and their historical behavior data can be updated preferentially to improve the control efficiency, can significantly suppress microgrid oscillations, improve system stability, is applicable to the microgrid coordinated control in the scenario of high new energy penetration rate, and has broad application prospects. Description of the Drawings

[0038] Figure 1 is the flowchart of the microgrid coordinated control method based on non-linear system behavior trajectory predictive control;

[0039] Figure 2It is a control structure diagram of a microgrid coordinated control method and an energy storage converter based on the prediction control of the behavior trajectory of a nonlinear system;

[0040] Figure 3 It is the topology diagram of the simulation system in the embodiment;

[0041] Figure 4 Taking the realization of oscillation suppression as an example, it is the effect diagram of the microgrid coordinated control method based on the prediction control of the behavior trajectory. Specific implementation manners

[0042] The technical solution of the present invention will be further elaborated in detail below in conjunction with the accompanying drawings of the specification and specific embodiments. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the technical field to which the present invention belongs. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention. In the following description, the expression "some embodiments" is described, which describes a subset of all possible embodiments. However, it should be understood that "some embodiments" can be the same subset or different subsets of all possible embodiments and can be combined with each other without conflict.

[0043] In the following description, a large number of specific details are given to provide a more thorough understanding of the present invention. However, it is obvious to those skilled in the art that the present invention can be implemented without one or more of these details. In other examples, in order to avoid confusion with the present invention, some technical features well known to the art are not described.

[0044] It should be understood that the present invention can be implemented in different forms and should not be construed as limited to the embodiments presented herein. On the contrary, providing these embodiments will make the disclosure thorough and complete and will fully convey the scope of the present invention to those skilled in the art. And the purpose of the terms used herein is only to describe specific embodiments and is not a limitation of the present invention. When used herein, the singular forms "a", "an" and "the" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the terms "comprising" and / or "including", when used in this specification, determine the presence of the described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or groups. When used herein, the term "and / or" includes any and all combinations of the related listed items.

[0045] It should be noted that when an element is referred to as "fixed to" another element, it can be directly on the other element or there may be an intermediate element. When an element is considered to be "connected" to another element, it can be directly connected to the other element or there may be an intermediate element at the same time. The terms "vertical", "horizontal", "inner", "outer", "left", "right" and similar expressions used herein are for illustrative purposes only and do not represent the only implementation.

[0046] To thoroughly understand the present invention, detailed structures will be presented in the following description to explain the technical solutions proposed by the present invention. The optional embodiments of the present invention are described in detail as follows. However, in addition to these detailed descriptions, the present invention may also have other implementations.

[0047] Please refer to the attached Figure 1 , this application provides a microgrid coordinated control method based on nonlinear system behavior trajectory prediction control. The control method includes the following steps:

[0048] S1. Excite the dynamic response of the microgrid unit by injecting an excitation signal, collect historical operation data and real-time operation data, and respectively construct a historical behavior description and a real-time behavior description characterizing the nonlinear behavior of the unit;

[0049] S2. Based on the historical behavior description and the real-time behavior description, generate an optimal control law by solving an optimization problem including the system stability objective, and superimpose the control instruction on the unit control loop;

[0050] S3. According to the dynamic harmonic characteristics of the unit connection point, screen the units that need to update the behavior description, and periodically update their historical behavior data to adapt to the dynamic changes of the system.

[0051] In an optional implementation, step S1 specifically includes continuously injecting full-band white noise into the control loop of the microgrid unit to excite the behavior trajectory data of the microgrid unit, obtaining the historical behavior description of the unit black box model, and collecting the operation state data of the microgrid unit in real time to obtain the real-time behavior description of the unit black box model.

[0052] Furthermore, before establishing a nonlinear behavior trajectory prediction model based on the historical behavior description and the real-time behavior description, it also includes constructing a new energy unit nonlinear system. Please refer to Figure 2 , Figure 2 is the control block diagram of a typical new energy unit, where C 1 is the DC side capacitor, L f , R f , C f are filter inductors, resistors and capacitors,L g is the inductance of the transmission line, i d,ref is the reference value of the current control loop, i d 、 i q are the dq-axis currents of the current control loop respectively, u d 、 u q are the voltage command values of the new energy unit control loop respectively. PI represents a proportional-integral controller. The constructed non-linear system of the new energy unit is described as:

[0053]

[0054] After linearizing the non-linear system of the new energy unit, it is described as:

[0055]

[0056] Among them, and are the vector fields of the new energy unit, represents the time instance of the continuous-time function of the new energy unit, represents the state sequence composed of data with length T, represents the input sequence composed of data with length T, represents the increment of the state sequence at time k, represents the increment of the input sequence at time k, represents the increment of the state sequence at time k + 1, represents the increment of the input sequence at time k + 1, represents under the sequences and the sequence the derivative of x with respect to f, represents under the sequences and the sequence the derivative of x with respect to h, represents under the sequences and the sequence the derivative of u with respect to f, represents under the sequences and the sequence the derivative of u with respect to h, where represents the state quantity of the non-linear system at time k, represents the output quantity of the non-linear system at time k, represents the increment of the output sequence at time k + 1.

[0057] Furthermore, the historical behavior description of the unit black-box model is obtained, specifically including: At At a moment, an input signal to the unit non - linear system inject a white noise signal with a per - unit value of noise power to trigger the persistent excitation of the output of the non - linear system Based on the sampling time sample and to obtain a non - linear historical measurement sequence of length where the non - linear historical measurement sequence includes an input sequence and an output sequence ;

[0058] where represents the input data at the i - th moment, and the input data simultaneously includes the d - axis current additional reference value signal of the converter and the q - axis current additional reference value signal , represents an input sequence composed of data of length T starting from the i - th moment, represents the output data at the i - th moment, represents an output sequence composed of data of length T starting from the i - th moment.

[0059] Furthermore, obtain descriptions of the historical behavior and real - time behavior of the non - linear system, specifically including:

[0060] Based on the input sequence and the output sequence, construct an input Hankel matrix and an output Hankel matrix with the same number of rows L and the number of columns T - L + 1 respectively. The j - th column of the Hankel matrix is the data from the k 0+ j -1 moment to the k 0+ L + j -2 moment, where, L ≤ T , j ≤ T - L +1. Taking the input Hankel matrix as an example:

[0061]

[0062] Horizontally divide both the input Hankel matrix and the output Hankel matrix into two parts. The first rows are used to implicitly estimate the initial state of the system, denoted by the subscript P, and the last N rows are used to predict the future trajectory of the system, denoted by the subscript F. The division result is:

[0063]

[0064] Among them, , , , are sub - matrices after the corresponding Hankel matrix is partitioned. The superscript represents the initial recording time of the data sequence, represents the input Hankel matrix, represents the output Hankel matrix;

[0065] The historical behavior data matrix at time is: , taking the historical behavior data matrix as the historical behavior of the nonlinear system;

[0066] Record the unit operation data starting from time k, and the time length of the recorded data is , the additional reference value signal of the d - axis current and the additional reference value signal of the q - axis current , the active power and the reactive power at the grid - connection point of the unit. Construct the real - time behavior data matrix at time k as: , where and are the input sequences starting from time k with a length of respectively, and are data sequences with a length of N to be predicted. Take the real - time behavior data matrix as the real - time behavior description.

[0067] Furthermore, the optimization problem is expressed as:

[0068]

[0069]

[0070] Among them, represents the input of the nonlinear system obtained by solving the optimization problem at time k, represents the output of the nonlinear system obtained by solving the optimization problem at time k, represents the vector composed of from time i to time i + j - 1, represents the vector composed of from time i to time i + j - 1, is the set value of the input of the nonlinear system at time k, is the set value of the output of the nonlinear system at time k. o represents a row vector, represents a column vector with a length of n + 1, is the control cost matrix, and are both output cost matrices, representing the output response value that the system should finally reach, is an auxiliary slack variable, and are regularization parameters, g is the decision variable of the above optimization problem, representing the Kronecker product.

[0071] In an optional embodiment, the solution of the above optimization problem is obtained by using a MATLAB solver or by programming to obtain the optimal control law , and the dq-axis current reference values of the converter are injected and , to achieve microgrid oscillation suppression.

[0072] Furthermore, after obtaining the latest output data of the microgrid unit, the real-time behavior data matrix is updated, and the real-time optimal control law is calculated again to achieve microgrid oscillation suppression.

[0073] Furthermore, according to the dynamic harmonic characteristics of the grid connection point of the unit, the units that need to update the behavior description are screened, specifically including: calculating the proportion of the harmonic content at the grid connection point of each new energy unit, and selecting the units for the timing update algorithm according to the cumulative sum threshold truncation method.

[0074] Among them, when calculating the proportion of the harmonic content at the grid connection point of each new energy unit, the grid connection point voltage of the new energy unit is measured in real time as x(m), and the signal is processed using a 4-term 5th-order Nuttall window function and Fourier decomposed as:

[0075]

[0076] where k FFT is the frequency index, X( k FFT ) is the signal at the frequency of k FFT after Fourier transform; is the window function, w is the variable of the window function; j is the imaginary unit; e is the natural base; M is the length of the voltage signal; FFT is the sampling point index of the voltage signal in the time domain;

[0077] For the i-th new energy unit, the proportion of the harmonic content at its grid connection point is calculated as:

[0078]

[0079] where, F all is the set of all frequency indices.

[0080] After calculating the proportion of harmonic content of the i-th new energy unit, sort the proportions of harmonic content of all new energy units from largest to smallest to obtain a sorted list. Cumulatively calculate the sorted proportions of harmonic content. When the cumulative calculation result ≥ 50%, select the units included in the cumulative calculation process as the set of units that need to update the algorithm. For example, if there are 7 units in the formed sorted list, after adding the proportions of harmonic content of the first 5 units, if the sum is greater than or equal to 50%, then select these 5 units as the units that need to be updated.

[0081] Furthermore, after a time of Tupdate, recalculate the optimal control law and the set of units that need to update the algorithm, where Tupdate is generally set to 1 minute.

[0082] According to some embodiments of the present invention, based on the above method, a microgrid oscillation suppression system is designed, which includes:

[0083] A non-linear system historical behavior data matrix construction module, which is used to collect the historical measurement sequences of the non-linear systems of the microgrid units under continuous excitation and construct a historical behavior data matrix;

[0084] A non-linear system real-time behavior data matrix construction module, which is used to collect the measurement sequences of the non-linear systems of the microgrid units in real time and establish a real-time behavior data matrix;

[0085] A microgrid coordinated control module, which is used to construct a behavior trajectory prediction controller suitable for microgrid coordinated control based on the historical behavior data matrix and the real-time behavior data matrix of the non-linear system, solve the optimization problem, and obtain the optimal control law to achieve microgrid oscillation suppression;

[0086] A unit update selection module, which is used to select the units suitable for the timed update algorithm based on the cumulative sum result of the harmonic content proportion at the grid connection point, and then update their historical behavior data matrix.

[0087] To verify the solution in this application, the method of the present invention was verified using the simulation model of the IEEE9-bus system. In the test system, some units in the original IEEE9-bus system were replaced with wind turbines and energy storage, as Figure 3 shown. In bus #1 of the test system, an energy storage with a rated power of 12 MW and a wind turbine with a rated power of 12 MW are connected. In bus #3, an energy storage with a rated power of 12 MW is connected. Among them, the converters are all grid-following control.

[0088] The effect diagram of the microgrid coordinated control method based on behavior trajectory prediction control is as Figure 4As shown in the figure above, the three-phase voltages of bus #1 are presented. The blue, yellow, and red curves represent the voltage curves of phase A, phase B, and phase C respectively; the figure below shows the three-phase voltages of bus #1, and the blue, yellow, and red curves represent the voltage curves of phase A, phase B, and phase C respectively. At the initial moment, 110Hz oscillation phenomena are detected in the three-phase voltage measurement signals of both bus #1 and bus #3 in the system. By comparing the active power curves of the system under different controls, for the period from 0 to 1.9 seconds, in the original system without any control, the active power output of the new energy oscillates continuously, and there is a risk of system instability; at 1.9 seconds, by adopting the behavior trajectory prediction control technology proposed in the present invention, the oscillation phenomena in the three-phase voltages are effectively suppressed.

[0089] The above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. The protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control, characterized in that: The control method comprises the following steps: By injecting excitation signals to stimulate the dynamic response of the microgrid unit, historical operation data and real-time operation data are collected to construct historical behavior description and real-time behavior description to characterize the nonlinear behavior of the unit respectively; Based on the historical behavior description and the real-time behavior description, the optimal control law is generated by solving the optimization problem including the system stability objective, and the control instructions are superimposed on the unit control loop; According to the dynamic harmonic characteristics of the unit grid connection point, select the units that need to update the behavior description, and periodically update their historical behavior data to adapt to the dynamic changes of the system; The dynamic response of the microgrid unit is stimulated by injecting excitation signals, including: At this moment, the input signal to the nonlinear system of the unit is The injected noise power per unit is White noise signal to trigger the nonlinear system output Continuous excitation based on the sampling time right , Sampling is performed to obtain a length of The nonlinear historical measurement sequence includes the input sequence and the output sequence ; in represents the input data at time i, and the input data also includes the d-axis current additional reference value signal of the converter and q-axis current additional reference value signal , represents the input sequence of data of length T starting at time i, Represents the output data at time i, Represents the output sequence consisting of data of length T starting at time i; According to the dynamic harmonic characteristics of the unit grid connection point, the units that need to update the behavior description are selected, including: calculating the proportion of harmonic content at the grid connection point of each new energy unit, and selecting the units for the timed update algorithm according to the cumulative and threshold truncation method; After calculating the harmonic content ratio of the i-th new energy unit, sort the harmonic content ratios of all new energy units from large to small to obtain a sorted list, and perform cumulative calculation on the sorted harmonic content ratios. When the cumulative calculation result is ≥50%, the units included in the cumulative calculation process are selected as the set of units that need to update the algorithm; When calculating the proportion of harmonic content at the grid connection point of each new energy unit, the voltage at the grid connection point of the new energy unit is measured in real time as x(m). The signal is processed using a 4-term 5th-order Nuttall window function and Fourier decomposition is performed as follows: in k FFT is the frequency index, X( k FFT ) is the frequency after Fourier transformation k FFT The signal of time; is the window function, w is the variable of the window function, j is the imaginary unit, e is the natural base, M is the length of the voltage signal, and FFT is the sampling point index of the voltage signal in the time domain.

2. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control according to claim 1, characterized in that: Based on the historical behavior description and the real-time behavior description, before establishing the nonlinear behavior trajectory prediction model, it also includes constructing a nonlinear system of new energy units, and describing the constructed nonlinear system of new energy units as follows: The nonlinear system of the new energy unit is linearized and described as: in, and is the vector domain of the new energy unit, represents the time instance of the continuous time function of the renewable energy unit, Represents a state sequence consisting of data of length T, Represents an input sequence consisting of data of length T, represents the increment of the state sequence at time k, represents the increment of the input sequence at time k, represents the increment of the state sequence at time k+1, represents the increment of the output sequence at time k+1, Indicates in sequence and sequence The derivative of x with respect to f is, Indicates in sequence and sequence The derivative of x with respect to h is, Indicates in sequence and sequence The derivative of u with respect to f is, Indicates in sequence and sequence The derivative of u with respect to h is, Represents the output of the nonlinear system at time k.

3. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control according to claim 2, characterized in that: Obtain historical and real-time descriptions of nonlinear system behavior, including: Based on the input sequence and the output sequence, construct an input Hankel matrix and an output Hankel matrix, each of which has L rows and T-L+1 columns; The input Hankel matrix and the output Hankel matrix are divided horizontally into two parts. The first N rows are used to implicitly estimate the initial state of the system, and the last N rows are used to predict the future trajectory of the system. The division results are: in, , , , is the submatrix after the corresponding Hankel matrix is ​​divided, and the superscript Indicates the initial recording time of the data sequence, represents the input Hankel matrix, Represents the output Hankel matrix; Historical behavior data matrix at each moment for: , taking the historical behavior data matrix as the historical behavior of the nonlinear system; Record the unit operation data starting at time k. The length of time for recording the data is , d-axis current additional reference value signal and q-axis current additional reference value signal , active power of the unit grid connection point and reactive power , construct the real-time behavior data matrix at time k as: ,in as well as They start at time k and have a length of The input sequence is as well as is a data sequence of length N to be predicted, and a real-time behavior data matrix is ​​used as the real-time behavior description.

4. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control according to claim 3, characterized in that: The optimization problem is expressed as: in, represents the nonlinear system input obtained by the optimization problem at time k, represents the nonlinear system output obtained by the optimization problem at time k, represents the time from time i to time i+j-1 The vector formed, represents the time from time i to time i+j-1 The vector formed, Enter the set value for the nonlinear system at time k, is the set value of the nonlinear system output at time k, o represents the row vector, represents a column vector of length n+1, is the control cost matrix, and are the output cost matrices, Indicates the output response value that the system should finally reach. is an auxiliary slack variable, and is the regularization parameter, g is the decision variable of the above optimization problem, represents the Kronecker product.

5. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control according to claim 4, characterized in that: Use MATLAB solver, or use programming to solve the optimization problem and obtain the optimal control law , and inject the converter dq axis current reference value and , realizing microgrid oscillation suppression.

6. A microgrid coordinated control method based on nonlinear system behavior trajectory predictive control according to claim 5, characterized in that: After Tupdate time has passed, the optimal control law and the set of units that need to update the algorithm are recalculated.

Images