Predictive control method and system for micro-grid system

Abstract

The invention discloses a prediction control method and system for a micro-grid system. The method comprises the steps of determining a state variable and a control signal of a simulation system model of a target micro-grid system; processing the state variable and the control signal through a Koopman operator theory, so as to construct a Koopman operator; performing optimization processing on the Koopman operator based on an extended dynamic mode decomposition technology to construct a micro-grid high-dimensional linear dynamic model corresponding to the target micro-grid system; according to a selected micro-grid expected condition, constructing a micro-grid prediction model in a high-dimensional space; based on the micro-grid high-dimensional linear dynamic model and the micro-grid prediction model, establishing a corresponding objective function and constraint conditions, and performing rolling optimization to obtain a global optimal solution; and realizing predictive control on the target microgrid system by using the input quantity corresponding to the globally optimal solution. According to the method provided by the embodiment of the invention, the modeling difficulty of the model can be reduced, the dynamic complexity of the system can be captured, and the control of the micro-grid system can be realized more accurately.

Description

Technical Field

[0001] The present invention relates to the field of power grid control technology, and in particular to a predictive control method and system for a microgrid system. Background Art

[0002] Microgrid refers to a micro power supply network based on distributed generation technology, mainly renewable energy, and using energy storage systems and control systems to achieve power balance within the network. Renewable energy is greatly affected by the environment and has the characteristics of randomness and intermittency. Nowadays, as more and more renewable energy and power electronic equipment are connected to the microgrid system, the stable operation and control of the microgrid system are challenged.

[0003] Model predictive control (MPC) is an effective way to solve this problem. However, in the existing technology, the nonlinearity of the system is not taken into account, resulting in poor modeling of the microgrid system. It is difficult to achieve efficient and accurate control using traditional control methods, which is not conducive to the stability of the microgrid system. Summary of the invention

[0004] The present invention provides a predictive control method and system for a microgrid system. The Koopman theory is used to improve the dimension of the microgrid model and map the nonlinear system to a high-dimensional linear space. This can not only reduce the difficulty of modeling, but also capture the dynamic complexity of the system, thereby more accurately controlling the microgrid system and ensuring the stability of the microgrid system.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides a predictive control method for a microgrid system, comprising:

[0006] Determine state variables and control signals of a simulation system model of a target microgrid system;

[0007] Processing the state variable and the control signal through Koopman operator theory to construct a Koopman operator;

[0008] The Koopman operator is optimized based on the extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory;

[0009] According to the selected microgrid expected conditions, a microgrid prediction model in high-dimensional space is constructed;

[0010] Based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, corresponding objective functions and constraints are established, and rolling optimization is performed to obtain a global optimal solution;

[0011] The predictive control of the target microgrid system is realized by using the input quantity corresponding to the global optimal solution.

[0012] As a preferred solution, the state variable includes one or more of the following: voltage, current, power, frequency and phase; and,

[0013] The control signal includes one or more of the following: a voltage regulation signal, a current regulation signal, a power regulation signal, a frequency regulation signal and a phase regulation signal.

[0014] As one preferred solution, the step of determining the state variables and control signals of the simulation system model of the target microgrid system includes:

[0015] Applying disturbance to the constructed simulation system model;

[0016] The disturbance results are analyzed according to the participating factor analysis method, and the state variable having the greatest influence on the simulation system model after the disturbance is selected as the control signal.

[0017] As one of the preferred solutions, the microgrid prediction model in high-dimensional space is constructed according to the selected microgrid expected conditions, including:

[0018] Determining the desired condition of the microgrid according to the load demand of the target microgrid system;

[0019] Acquire historical load change data of the target microgrid system;

[0020] The historical load change data is trained using a deep learning algorithm to construct the microgrid prediction model in a high-dimensional space.

[0021] As one of the preferred solutions, the predictive control of the target microgrid system is realized by using the input quantity corresponding to the global optimal solution, including:

[0022] Inputting the input quantity corresponding to the global optimal solution into the target microgrid system as a control signal, wherein the input quantity at least includes the output power of the power generation equipment and / or the charging and discharging power of the energy storage system;

[0023] The operating status of the target microgrid system is monitored in real time, and the microgrid prediction model is optimized and adjusted according to the operating status.

[0024] Another embodiment of the present invention provides a predictive control system for a microgrid system, comprising:

[0025] A determination module, used to determine state variables and control signals of a simulation system model of a target microgrid system;

[0026] A Koopman operator module, used for processing the state variable and the control signal through Koopman operator theory to construct a Koopman operator;

[0027] A high-dimensional module, used for optimizing the Koopman operator based on the extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory;

[0028] A prediction module is used to construct a microgrid prediction model in a high-dimensional space according to the selected microgrid expected conditions;

[0029] A solution module, used to establish corresponding objective functions and constraints based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, and to perform rolling optimization to obtain a global optimal solution;

[0030] A control module is used to implement predictive control of the target microgrid system with an input quantity corresponding to the global optimal solution.

[0031] As a preferred solution, the state variable includes one or more of the following: voltage, current, power, frequency and phase; and,

[0032] The control signal includes one or more of the following: a voltage regulation signal, a current regulation signal, a power regulation signal, a frequency regulation signal and a phase regulation signal.

[0033] As one preferred solution, the determining module includes:

[0034] A disturbance unit, used for applying disturbance to the constructed simulation system model;

[0035] The factor unit is used to analyze the disturbance result according to the participating factor analysis method, and select the state variable that has the greatest impact on the simulation system model after the disturbance as the control signal.

[0036] As one preferred solution, the prediction module includes:

[0037] A load unit, configured to determine the desired condition of the microgrid according to the load demand of the target microgrid system;

[0038] A history unit, used to obtain historical load change data of the target microgrid system;

[0039] A construction unit is used to train the historical load change data using a deep learning algorithm to construct the microgrid prediction model in a high-dimensional space.

[0040] As one preferred solution, the control module includes:

[0041] An input unit, used to input an input quantity corresponding to the global optimal solution into the target microgrid system as a control signal, wherein the input quantity at least includes an output power of a power generation device and / or a charge and discharge power of an energy storage system;

[0042] The monitoring unit is used to monitor the operating status of the target microgrid system in real time and optimize and adjust the microgrid prediction model according to the operating status.

[0043] Compared with the prior art, the embodiments of the present invention have the following advantages:

[0044] (1) Through the Koopman theory, this scheme can improve the dimension of the microgrid model and map the nonlinear system into a high-dimensional linear space. This can not only reduce the difficulty of modeling, but also capture the complexity of system dynamics and better describe the entire dynamic behavior of the actual power grid system.

[0045] (2) This scheme uses the Koopman operator to establish a finite high-dimensional linear dynamic model of the microgrid. On this basis, combined with model predictive control, the system behavior can be directly learned from the collected microgrid data. In addition, since the controller uses a high-dimensional linear model, it avoids non-convex optimization problems, thereby reducing computational complexity and improving computational efficiency, providing technical support for the safe and stable operation of the microgrid. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 is a flow chart of a predictive control method for a microgrid system in one embodiment of the present invention;

[0047] Figure 2 is a schematic diagram of a control block diagram in one embodiment of the present invention;

[0048] Figure 3 It is a structural schematic diagram of a predictive control system of a microgrid system in one embodiment of the present invention.

[0049] Reference numerals:

[0050] Among them, 11, determination module; 12, Koopman operator module; 13, high-dimensional module; 14, prediction module; 15, solution module; 16, control module. DETAILED DESCRIPTION

[0051] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0052] In the description of this application, the terms "first", "second", "third", etc. are used for descriptive purposes only and are not to be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first", "second", "third", etc. may explicitly or implicitly include one or more of the feature. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0053] In the description of the present application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be a connection between the two elements. The terms "vertical", "horizontal", "left", "right", "upper", "lower" and similar expressions used herein are only for illustrative purposes, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. The term "and / or" used herein includes any and all combinations of one or more related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0054] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in the present invention have the same meaning as those commonly understood by those skilled in the art. The terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. For those of ordinary skill in the art, the specific meanings of the above terms in this application can be understood by specific circumstances.

[0055] An embodiment of the present invention provides a predictive control method for a microgrid system. For details, see Figure 1 , Figure 1 The flowchart of the predictive control method of the microgrid system in one embodiment of the present invention is shown, which includes steps S1 to S6:

[0056] S1. Determine the state variables and control signals of the simulation system model of the target microgrid system;

[0057] S2. Processing the state variable and the control signal by Koopman operator theory to construct a Koopman operator;

[0058] S3. Optimizing the Koopman operator based on extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory;

[0059] S4. According to the selected microgrid expected conditions, a microgrid prediction model in high-dimensional space is constructed;

[0060] S5. Based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, corresponding objective functions and constraints are established, and rolling optimization is performed to obtain a global optimal solution;

[0061] S6. Implement predictive control of the target microgrid system using the input quantity corresponding to the global optimal solution.

[0062] The embodiment of the present invention uses the Koopman operator to establish a finite high-dimensional linear dynamic model of the microgrid. On this basis, combined with model predictive control, a microgrid predictive stability control method based on the Koopman operator is proposed. This method can directly learn the behavior of the system from the collected microgrid data, and because the controller adopts a high-dimensional linear model, it avoids non-convex optimization problems, thereby reducing the computational complexity and improving the computational efficiency, providing technical support for achieving safe and stable operation of the microgrid. In the above embodiment, the state variables are key parameters that describe the dynamic behavior of the system, and they can reflect the current state of the system and future change trends. In the microgrid system, common state variables include: voltage, current, power, frequency and phase; the control signal is an input used to adjust the system state to achieve the control target. In the microgrid system, common control signals include: voltage regulation signal, current regulation signal, power regulation signal, frequency regulation signal and phase regulation signal. When selecting the control signal, factors such as the control target of the microgrid, the stability of the system, the characteristics of the equipment and the simulation requirements need to be considered. For example, in an isolated microgrid, it may be necessary to maintain system stability by adjusting voltage and frequency; in a grid-connected microgrid, it may be necessary to adjust power and phase to achieve coordinated operation with the main grid, which is determined by actual needs and will not be further elaborated in the embodiments of the present invention.

[0063] In the above embodiment, a disturbance is applied to the constructed simulation system model, and the result after the disturbance is analyzed according to the participation factor analysis method. The participation factor analysis method is used to select the control signal. By selecting the state variable that has the greatest impact on the microgrid after the disturbance as the control signal, the future state of the system can be predicted more accurately, and the corresponding control strategy can be formulated to optimize the performance of the system. This control signal selection method based on the participation factor analysis method not only improves the accuracy of predictive control, but also helps to achieve stable operation and efficient management of the microgrid.

[0064] Further, in the above embodiment, constructing a microgrid prediction model in a high-dimensional space includes: selecting the desired conditions of the microgrid, and in this embodiment, the selection is preferably made based on the load demand of the microgrid. Of course, the selection can also be made based on the forecast of renewable energy generation or the state of the energy storage system, etc., which is not specifically limited in the embodiment of the present invention. A high-dimensional prediction model of the microgrid is constructed using machine learning or deep learning algorithms (such as time series prediction models, neural networks, etc.), which should be able to predict the state of the microgrid in the future based on current and historical data.

[0065] The process of the embodiment of the present invention is described in detail below. The selection of state variables and control signals has been described above and will not be repeated here.

[0066] In an embodiment of the present invention, a finite high-dimensional linear dynamic model of a microgrid is established using the Koopman operator to transform the nonlinear description of the system into a linear representation in a high-dimensional space. The high-dimensional linearization model can more accurately capture the nonlinear characteristics of the microgrid and better describe the entire dynamic behavior of the actual engineering system.

[0067] Specifically, consider a nonlinear dynamic system:

[0068] x(k+1)=f(x(k),u(k)) (1)

[0069] Definition The Koopman operator κ is a linear operator acting on the observation function g(x).

[0070] κg(x(k))=g(x(k+1))=g(f(x(k),u(k))) (2)

[0071] Since the Koopman operator is infinite-dimensional, it is necessary to obtain a finite-dimensional value in practical applications. The finite-dimensional approximation of the Koopman operator is solved based on the Extended Dynamic Mode Decopmosition (EDMD) to analyze the dynamic behavior of the system. The lifting state is:

[0072]

[0073] in, is the basis function, where i = 1, 2, ..., β; β is the dimension of the system in high-dimensional space.

[0074] The basis function in formula (3) can generally be selected as thin plate spline radial basis function, Gaussian function, inverse multi-quadratic function, etc.

[0075] The constructed Koopman operator high-dimensional linear model is:

[0076]

[0077] In the formula, is the state of the high-dimensional system, β is the dimension after dimensionality increase, z(k)=[z1(k)z2(k)...z β (k)] T , is an estimate of the state of the original nonlinear system. are all linear constant matrices, m is the number of columns of matrix B, and n is the number of rows of matrix C.

[0078] For the actual microgrid model, the state variables and input data of the system are collected based on the simulation experiment, and the data set is defined as follows:

[0079]

[0080] When Thin Plate Spline Radial Basis Function (RBF) is selected:

[0081]

[0082] Among them, c i is the center point of the i-th basis function. If the center point is [i,...,i] T but:

[0083]

[0084] Therefore, formula (3) can be transformed into:

[0085]

[0086] Datasets X and Y are raised to X through basis functions lift and Y lift :

[0087]

[0088] According to the EDMD method, A, B, and C in formula (4) solve the following optimization problem:

[0089]

[0090] The solution of formula (10) is:

[0091]

[0092] in, is the Moore-Penrose pseudoinverse.

[0093] Substituting A, B, and C obtained in equation (11) into equation (4), we can obtain a high-dimensional linear dynamic model based on the Koopman operator.

[0094] In the above embodiment, the state space model of the microgrid can be expressed as:

[0095]

[0096] Discretizing equation (12) yields:

[0097]

[0098] Among them, x(k) is the state variable of the system at time k, and u(k) is the input variable of the system at time k.

[0099] According to the basic principle of model predictive control, the future state variables of the microgrid system are predicted with the state variable x(k) at the current time k as the starting point:

[0100]

[0101] In formula (14), x(k+i|k) is the state variable at the k+ith moment predicted at the kth moment, i=1,2,...,N, and u(k+j|k) is the control input at the k+jth moment predicted at the kth moment, j=1,2,...,N-1.

[0102] Substituting the predicted state variables obtained by equation (14) into equation (13), the future output of the microgrid system can be obtained:

[0103]

[0104] In formula (15), Y is the microgrid output, U(k) = [u(k|k) u(k+1|k) u(k+2|k) … u(k+N-1|k)] T

[0105] The Koopman model of the microgrid is shown in formula (4), and the prediction of the future state of the high-dimensional system z(k) can be written as:

[0106]

[0107] Where N C To control the time domain, N P To predict the time domain, the output of the system is defined as y(k), that is, yk()x=k(). According to formula (4), the predicted output is expressed as:

[0108]

[0109] Based on formula (15), the prediction equation of formula (17) can be rewritten as:

[0110]

[0111] in,

[0112] The reference trajectory is usually expressed in the form of a time series, containing the expected value of the system state or output at each time step. The reference trajectory can be expressed as:

[0113]

[0114] Among them, y ref (k+i) is the expected state or output value at the i-th time step.

[0115] Combined with the prediction model of the system and given the expected output of the system, the objective function and corresponding constraints of the entire system at the current k moment can be established:

[0116]

[0117] In formula (20), Y is the predicted output of the system, Y ref is the reference output of the system, Q is the error weighting matrix, and R is the control input weighting matrix.

[0118] It can be seen from formula (20) that the objective function is a standard quadratic programming form. By solving the quadratic programming problem, the optimal input sequence at the current moment is obtained, and the first optimization result u(k|k) is used as the input of the system to obtain the system output at time k+1. Repeated iterations and rolling optimization can eventually obtain the global optimal solution of the entire system, so that the system output reaches the expected value.

[0119] Substituting formula (18) into formula (20), we can get the objective function of the microgrid model:

[0120]

[0121] And meet:

[0122]

[0123] Among them: Q is the state weight matrix of the nonlinear system; R is the weight matrix of the control input; F is the terminal penalty matrix, all of which are positive definite weight matrices.

[0124] In order to improve the solving efficiency, the performance index is transformed into:

[0125]

[0126] In formula (23):

[0127]

[0128] Formula (21) can be solved as a standard linear quadratic programming problem, so that the global optimal solution can be quickly obtained. By solving formula (21), we can get N in the prediction time domain: P The optimal control sequence U(k) in .

[0129] For details, see Figure 2 , Figure 2 The schematic diagram of the control block diagram in one embodiment of the present invention is shown. Of course, in practical applications, it is necessary to adjust the Q and R matrices so that the system can achieve high-precision trajectory tracking within the specified constraints.

[0130] In the above embodiment, in order to achieve predictive control of the target microgrid system, the input quantity corresponding to the global optimal solution (such as the output power of the power generation equipment, the charging and discharging strategy of the energy storage system, etc.) needs to be input into the microgrid system as a control signal. Then the operating status of the microgrid is monitored in real time, and the prediction model and optimization algorithm are adjusted according to the actual situation to ensure the accuracy and effectiveness of the predictive control.

[0131] Another embodiment of the present invention provides a predictive control system for a microgrid system. For details, see Figure 3 , Figure 3 A schematic diagram of a predictive control system for a microgrid system in one embodiment of the present invention is shown, comprising:

[0132] A determination module 11 is used to determine the state variables and control signals of the simulation system model of the target microgrid system;

[0133] A Koopman operator module 12, used for processing the state variable and the control signal through Koopman operator theory to construct a Koopman operator;

[0134] A high-dimensional module 13 is used to optimize the Koopman operator based on the extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory;

[0135] A prediction module 14 is used to construct a microgrid prediction model in a high-dimensional space according to the selected microgrid expected conditions;

[0136] A solution module 15 is used to establish corresponding objective functions and constraints based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, and to obtain a global optimal solution through rolling optimization;

[0137] The control module 16 is used to implement predictive control of the target microgrid system with the input quantity corresponding to the global optimal solution.

[0138] Further, in the above embodiment, the state variable includes one or more of the following: voltage, current, power, frequency and phase; and,

[0139] The control signal includes one or more of the following: a voltage regulation signal, a current regulation signal, a power regulation signal, a frequency regulation signal and a phase regulation signal.

[0140] Furthermore, in the above embodiment, the determining module includes:

[0141] A disturbance unit, used for applying disturbance to the constructed simulation system model;

[0142] The factor unit is used to analyze the disturbance result according to the participating factor analysis method, and select the state variable that has the greatest impact on the simulation system model after the disturbance as the control signal.

[0143] Furthermore, in the above embodiment, the prediction module includes:

[0144] A load unit, configured to determine the desired condition of the microgrid according to the load demand of the target microgrid system;

[0145] A history unit, used to obtain historical load change data of the target microgrid system;

[0146] A construction unit is used to train the historical load change data using a deep learning algorithm to construct the microgrid prediction model in a high-dimensional space.

[0147] Furthermore, in the above embodiment, the control module includes:

[0148] An input unit, used to input an input quantity corresponding to the global optimal solution into the target microgrid system as a control signal, wherein the input quantity at least includes an output power of a power generation device and / or a charge and discharge power of an energy storage system;

[0149] The monitoring unit is used to monitor the operating status of the target microgrid system in real time and optimize and adjust the microgrid prediction model according to the operating status.

[0150] The predictive control method and system for a microgrid system provided by an embodiment of the present invention have the beneficial effects of at least one of the following:

[0151] (1) Through the Koopman theory, this scheme can improve the dimension of the microgrid model and map the nonlinear system into a high-dimensional linear space. This can not only reduce the difficulty of modeling, but also capture the complexity of system dynamics and better describe the entire dynamic behavior of the actual power grid system.

[0152] (2) This scheme uses the Koopman operator to establish a finite high-dimensional linear dynamic model of the microgrid. On this basis, combined with model predictive control, the system behavior can be directly learned from the collected microgrid data. In addition, since the controller uses a high-dimensional linear model, it avoids non-convex optimization problems, thereby reducing computational complexity and improving computational efficiency, providing technical support for the safe and stable operation of the microgrid.

[0153] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A predictive control method for a microgrid system, characterized in that: include: Determine state variables and control signals of a simulation system model of a target microgrid system; Processing the state variable and the control signal through Koopman operator theory to construct a Koopman operator; The Koopman operator is optimized based on the extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory; According to the selected microgrid expected conditions, a microgrid prediction model in high-dimensional space is constructed; Based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, corresponding objective functions and constraints are established, and rolling optimization is performed to obtain a global optimal solution; The predictive control of the target microgrid system is realized by using the input quantity corresponding to the global optimal solution.

2. The predictive control method for a microgrid system according to claim 1, characterized in that: The state variables include one or more of the following: voltage, current, power, frequency and phase; and, The control signal includes one or more of the following: a voltage regulation signal, a current regulation signal, a power regulation signal, a frequency regulation signal and a phase regulation signal.

3. The predictive control method for a microgrid system according to claim 1, characterized in that: The step of determining the state variables and control signals of the simulation system model of the target microgrid system includes: Applying disturbance to the constructed simulation system model; The disturbance results are analyzed according to the participating factor analysis method, and the state variable having the greatest influence on the simulation system model after the disturbance is selected as the control signal.

4. The predictive control method for a microgrid system according to claim 1, characterized in that: The microgrid prediction model in high-dimensional space is constructed according to the selected microgrid expected conditions, including: Determining the desired condition of the microgrid according to the load demand of the target microgrid system; Acquire historical load change data of the target microgrid system; The historical load change data is trained using a deep learning algorithm to construct the microgrid prediction model in a high-dimensional space.

5. The predictive control method for a microgrid system according to claim 1, characterized in that: The predictive control of the target microgrid system is realized by using the input quantity corresponding to the global optimal solution, including: Inputting the input quantity corresponding to the global optimal solution into the target microgrid system as a control signal, wherein the input quantity at least includes the output power of the power generation equipment and / or the charging and discharging power of the energy storage system; The operating status of the target microgrid system is monitored in real time, and the microgrid prediction model is optimized and adjusted according to the operating status.

6. A predictive control system for a microgrid system, characterized in that: include: A determination module, used to determine state variables and control signals of a simulation system model of a target microgrid system; A Koopman operator module, used for processing the state variable and the control signal through Koopman operator theory to construct a Koopman operator; A high-dimensional module, used for optimizing the Koopman operator based on the extended dynamic mode decomposition technology to construct a microgrid high-dimensional linear dynamic model corresponding to the target microgrid system, wherein the high-dimensional linear dynamic equations in the microgrid high-dimensional linear dynamic model are constructed by Koopman operator theory; A prediction module is used to construct a microgrid prediction model in a high-dimensional space according to the selected microgrid expected conditions; A solution module, used to establish corresponding objective functions and constraints based on the microgrid high-dimensional linear dynamic model and the microgrid prediction model, and to perform rolling optimization to obtain a global optimal solution; A control module is used to implement predictive control of the target microgrid system with an input quantity corresponding to the global optimal solution.

7. The predictive control system of the microgrid system according to claim 6, characterized in that: The state variables include one or more of the following: voltage, current, power, frequency and phase; and, The control signal includes one or more of the following: a voltage regulation signal, a current regulation signal, a power regulation signal, a frequency regulation signal and a phase regulation signal.

8. The predictive control system of the microgrid system according to claim 6, characterized in that: The determination module comprises: A disturbance unit, used for applying disturbance to the constructed simulation system model; The factor unit is used to analyze the disturbance result according to the participating factor analysis method, and select the state variable that has the greatest impact on the simulation system model after the disturbance as the control signal.

9. The predictive control system of the microgrid system according to claim 6, characterized in that: The prediction module comprises: A load unit, configured to determine the desired condition of the microgrid according to the load demand of the target microgrid system; A history unit, used to obtain historical load change data of the target microgrid system; A construction unit is used to train the historical load change data using a deep learning algorithm to construct the microgrid prediction model in a high-dimensional space.

10. The predictive control system of the microgrid system according to claim 6, characterized in that: The control module comprises: An input unit, used to input an input quantity corresponding to the global optimal solution into the target microgrid system as a control signal, wherein the input quantity at least includes an output power of a power generation device and / or a charge and discharge power of an energy storage system; The monitoring unit is used to monitor the operating status of the target microgrid system in real time and optimize and adjust the microgrid prediction model according to the operating status.

Images