Direct-current micro-grid distributed all-drive prediction method and system based on autoregression model

Abstract

The invention provides a direct-current micro-grid distributed all-drive prediction method and system based on an autoregression model. The method comprises the steps of performing decoupling analysis on a direct-current micro-grid coupling model to obtain an all-drive decoupling prediction model; an output current in a prediction time domain is obtained based on an all-drive decoupling prediction model, and on the basis of a control quantity in a control time domain predicted at the last moment, an autoregression model is adopted to further predict a required control quantity prediction value which cannot be directly obtained, and a prediction value is obtained; according to the obtained predicted value, designing a loss function and solving the obtained control quantity; designing a voltage amplitude limiter according to the obtained final control quantity, and ensuring that a bus voltage reference value fluctuates in an allowable range; and inputting the bus voltage reference value corresponding to the voltage amplitude limiter into a voltage and current double-loop controller, so that the bus voltage tracks the bus voltage reference value.

Description

Technical Field

[0001] The present invention belongs to the technical field of distributed control of DC microgrids, and in particular relates to a distributed full-drive prediction method and system for DC microgrids based on an autoregressive model. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] A DC microgrid is a local power network based on direct current (DC) distribution that can operate independently or in conjunction with the alternating current (AC) main grid, providing efficient and flexible power management for distributed energy resources, energy storage systems, and loads.

[0004] As a key component of distributed energy systems, DC microgrids have the advantages of flexible structure, high energy conversion efficiency, and easy integration of renewable energy. They have broad application prospects in smart power distribution, electric vehicle charging stations, and off-grid power supply.

[0005] The mathematical model of a traditional DC microgrid typically consists of multiple key components, including distributed power sources (such as photovoltaics and fuel cells), energy storage systems (such as batteries), loads, and power electronic converters. These models often exhibit a high degree of nonlinearity and coupling, manifesting as a strong dynamic correlation between bus voltage and power distribution, significantly increasing the complexity of controller design. This coupling makes it difficult for conventional control strategies (such as droop control and voltage-current dual-loop control) to achieve precise current sharing while maintaining voltage stability. This is particularly true when faced with sudden load changes or fluctuating distributed power output, which can severely impact system dynamic performance. Existing approaches often rely on complex decoupling compensation or hierarchical control architectures, which not only increase hardware implementation costs but also reduce system response speed and robustness.

[0006] In order to reduce system coupling, the existing technology proposes a partial decoupling model (such as a simplified model based on power-voltage decoupling), but it ignores the high-order dynamic characteristics and still has current sharing errors during transient processes.

[0007] Fully Actuated System Theory (FAS) is a multivariable control-based approach that aims to achieve dynamic decoupling and precise control of the system by designing completely independent control channels. In DC microgrids, FAS can be applied to construct a decoupling model, independently adjusting the dynamic behavior of each subsystem, such as voltage, power, and energy storage status, to improve system stability and response speed. In contrast, the decoupling model constructed using FAS can precisely achieve input-output decoupling through state feedback, providing a superior framework for control design.

[0008] However, predictive control methods based on all-wheel drive models have significant drawbacks in existing applications: their rolling optimization time domain typically covers only a limited control step size, resulting in the neglect of dynamic changes in the predicted control variable outside the time domain. This problem is particularly prominent in scenarios with large dynamic differences in distributed power sources or communication delays, manifesting as bus voltage oscillations and reduced current sharing accuracy. Summary of the Invention

[0009] To overcome the above-mentioned deficiencies of the prior art, the present invention provides a distributed full-drive prediction method for a DC microgrid based on an autoregressive model, in which each distributed converter can still achieve current proportional equalization and voltage stability within an allowable range.

[0010] To achieve the above objectives, one or more embodiments of the present invention provide the following technical solutions:

[0011] In the first aspect, a distributed full-drive prediction method for a DC microgrid based on an autoregressive model is disclosed, comprising:

[0012] Decoupling analysis of the DC microgrid coupling model is performed to obtain an all-wheel drive decoupling prediction model;

[0013] The output current in the prediction time domain is obtained based on the all-wheel drive decoupling prediction model. Based on the control quantity in the control time domain predicted at the previous moment, the autoregressive model is used to further predict the required control quantity prediction value that cannot be directly obtained to obtain the predicted value;

[0014] According to the obtained predicted value, the loss function is designed and the obtained control quantity is solved;

[0015] Design a voltage limiter based on the final control quantity obtained to ensure that the bus voltage reference value fluctuates within the allowable range;

[0016] The bus voltage reference value corresponding to the above voltage limiter is input into the voltage-current dual-loop controller to make the bus voltage track the bus voltage reference value.

[0017] As a further technical solution, the DC microgrid coupling model includes the following steps during construction:

[0018] Constructing a first matrix including: a matrix of output voltage, output current, difference between rated voltage and output voltage, and difference between rated current and output current;

[0019] The second matrix is constructed including: voltage source voltage, current loop proportional coefficient*voltage loop proportional coefficient / inductance, and a matrix of voltage loop proportional coefficient.

[0020] As a further technical solution, a full-drive method is used to convert the DC microgrid coupling model into a full-drive decoupling prediction model.

[0021] As a further technical solution, the autoregressive prediction model is constructed by considering: the control time domain of the control quantity, the prediction time domain of the output current, the constant coefficient, the difference between the predicted value and the true value, and the proportional coefficient.

[0022] As a further technical solution, the voltage limiter is as follows:

[0023]

[0024] in Represents the voltage deviation value allowed by the system, Represents the bus voltage reference value of the system input.

[0025] As a further technical solution, the invention further includes: performing distributed control by a controller provided at each converter:

[0026] The controller set at each converter is to build a DC microgrid topology containing multiple converters, set a voltage and current dual-loop control and a controller of the proposed control strategy at each converter, and connect each converter to a distributed communication network.

[0027] Secondly, a distributed full-drive prediction system for DC microgrids based on an autoregressive model is disclosed, including:

[0028] The all-wheel drive model prediction module is configured to: perform decoupling analysis on the DC microgrid coupling model to obtain an all-wheel drive decoupling prediction model, and obtain an output current in a prediction time domain based on the all-wheel drive decoupling prediction model;

[0029] The autoregressive model prediction module is configured to: based on the control quantity within the control time domain predicted at the previous moment, use the autoregressive model to further predict the required control quantity prediction value that cannot be directly obtained to obtain the predicted value;

[0030] The predictive control solution module is configured to: design a loss function based on the obtained predicted value and the predicted output current and solve the obtained control variable;

[0031] The voltage limiting module is configured to: design a voltage limiter according to the obtained final control quantity to ensure that the bus voltage reference value fluctuates within an allowable range;

[0032] The microgrid control module is configured to: input the bus voltage reference value corresponding to the above-mentioned voltage limiter into the voltage-current dual-loop controller to make the bus voltage track the bus voltage reference value.

[0033] One or more of the above technical solutions have the following beneficial effects:

[0034] The technical solution of the present invention provides a distributed full-drive predictive control technology for a DC microgrid based on an autoregressive model. By simplifying and decoupling the traditional DC microgrid coupling model, a full-drive decoupling prediction model is obtained. Subsequently, based on the predicted control quantity at the previous moment, the autoregressive model is used to predict the control quantity prediction value outside the control time domain, and based on the obtained full-drive decoupling prediction model, the output current prediction value in the prediction time domain is obtained. Based on the obtained prediction quantity, a loss function is designed and solved to obtain the control quantity that satisfies the system current proportional distribution. A voltage limiter is designed to further correct the control quantity to ensure that the bus voltage fluctuates within an allowable range. The present invention uses an autoregressive prediction model to predict the control quantity beyond the time domain, thereby achieving a more accurate rolling optimization process, integrating a new control strategy for long-time domain prediction, and has important engineering value for improving the stability and current sharing performance of the DC microgrid.

[0035] Advantages of additional aspects of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] The accompanying drawings, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0037] Figure 1 This is a flow chart of the method of Example 1;

[0038] Figure 2 This is a schematic diagram of the DC microgrid structure and control method of Example 1;

[0039] Figure 3 This is a schematic diagram of a microgrid for simulation verification of Example 1;

[0040] Figure 4 The results of the resistive load step change in Example 1 are shown in FIG. 1 , wherein (a) is the output current curve; (b) is the output voltage curve;

[0041] Figure 5 Graphs showing the results of a constant power load step change in Example 1; (a) is the output current curve; (b) is the output voltage curve. DETAILED DESCRIPTION

[0042] It should be noted that the following detailed descriptions are exemplary and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which the present invention belongs.

[0043] It should be noted that the terms used herein are for describing particular embodiments only and are not intended to limit the exemplary embodiments according to the present invention.

[0044] In the absence of conflict, the embodiments of the present invention and the features thereof may be combined with each other.

[0045] Aiming at the out-of-time dynamic optimization problem in the predictive control of the all-wheel drive decoupling model, an autoregressive model is used to perform long-time domain prediction of the control variables beyond the control time domain. A new control strategy integrating long-time domain prediction is proposed, which has important engineering value for improving the stability and current sharing performance of the DC microgrid.

[0046] Example 1

[0047] This embodiment discloses a distributed full-drive prediction method for a DC microgrid based on an autoregressive model, including:

[0048] Step S1: A traditional coupling model of a DC microgrid is given, and a full-drive method is used to perform decoupling analysis on it, and finally a full-drive decoupling prediction model is obtained.

[0049] According to the attached Figure 2 The circuit system structure shown in the figure takes into account the droop control and the proportional-integral dual-loop control structure, and gives its traditional model. The traditional model of the DC microgrid is expressed as follows:

[0050]

[0051] y i =f i (x i )

[0052] where f i (x i )=G i x i , G i =

[0100] . The corresponding matrix represents the output voltage, output current, the difference between the rated voltage and the output voltage, and the difference between the rated current and the output current. The corresponding matrix represents 0, voltage source voltage, current loop proportional coefficient * voltage loop proportional coefficient / inductance, voltage loop proportional coefficient, u i Represents the control signal input, and its design process will be given later.

[0053] and

[0054]

[0055] Among them, C i Represents capacitance, R i Represents the load, Rij Represents the line resistance between adjacent power generation units, L i represents the line inductance, Represents the local line resistance. Represent the current loop proportional coefficient and integral coefficient respectively. Respectively represent the voltage loop proportional coefficient and integral coefficient, h i Represents the corresponding droop coefficient. For detailed physical distribution of parameters, see the attached Figure 2 .

[0056]

[0057] The full-wheel drive method is used to convert it into a full-wheel drive system model, decouple the DC microgrid model mentioned above, and achieve the current distribution goal in a simpler way. The current state in state x is decomposed and the corresponding full-wheel drive system model is obtained, which is expressed as follows:

[0058]

[0059] in, is f i To W i Lie derivative of .

[0060]

[0061] Therefore, the system input, i.e. the final control quantity, can be expressed as

[0062]

[0063] in The control input designed for the subsequent control strategy is set by setting the first item in the brackets as follows

[0064]

[0065] Therefore, the all-wheel drive decoupling prediction model can be finally expressed as follows

[0066]

[0067] Among them I i is the output current on bus i, k represents the discrete time, T s is the system sampling step size. Represents the local control quantity designed based on predictive control.

[0068] Step S2: Based on the control quantity prediction value at the previous moment, the above-mentioned all-wheel drive decoupling prediction model and the autoregressive model are used to predict the output current and control quantity at the current moment.

[0069] According to the all-wheel drive decoupling prediction model, the current at time k+n is obtained from k+n-1, so the corresponding output current prediction can be achieved. The specific predicted output current is as follows:

[0070]

[0071] in,

[0072]

[0073] The autoregressive prediction model is used to predict the control value outside the control time domain in the algorithm. For estimation, the autoregressive forecasting model can be designed as:

[0074]

[0075] Among them, N u Represents the control time domain of the control quantity, N y represents the predicted time domain of the output current, Represents that at time k, k+N u The predicted value of the step. v represents the constant coefficient. θ represents the difference between the predicted value and the true value, and κ is the corresponding proportional coefficient.

[0076] Step S3: Based on the obtained predicted value, design a loss function and solve the obtained control quantity.

[0077] Based on the control variable prediction value at the previous moment, the output current prediction value at the current moment, and the unknown control variable prediction value estimated by the autoregressive model at the current moment, to ensure that the obtained control law can achieve the final consistency of the weighted current, the cost function is designed as follows:

[0078]

[0079] where a ij is the communication weight, η i >0 represents a punishment factor.

[0080] make Available

[0081]

[0082] in and

[0083] Finally, the simplified form of the required control quantity is as follows:

[0084]

[0085] Where Λ is a matrix satisfying: and For any i≠j, Ψ is a matrix satisfying: and For any i≠j. s i =[0…0 1].

[0086] according to The final system input is as follows

[0087]

[0088] Step S4: Based on the obtained final control quantity, i.e., the system input, a voltage limiter is designed to ensure that the bus voltage reference value fluctuates within an allowable range.

[0089] The voltage limiter is as follows:

[0090]

[0091] in Represents the voltage deviation value allowed by the system, Represents the bus voltage reference value of the system input. Represents the expected rated value of the system voltage, V i is the actual measured voltage output value.

[0092] Step S5: Input the bus voltage reference value into the voltage-current dual-loop controller to ensure that the bus voltage tracks the reference value, thereby achieving the goals of current proportional distribution and voltage fluctuation within the allowable range.

[0093] This reference value is the result of the control quantity output of the proposed control strategy after being corrected by the voltage limiter. It has the ability to ensure that the weighted current of each distributed unit tends to be consistent and to correct the voltage, thereby achieving the goal of current proportional distribution and voltage fluctuation within the allowable range.

[0094] As an example, a distributed all-wheel drive predictive control method for a DC microgrid based on an autoregressive model ensures that even with varying resistive loads, each distributed converter can maintain current proportionality and maintain voltage stability within an acceptable range. Specifically, under varying resistive and constant power loads, the proposed control strategy maintains consistency in the weighted currents of each distributed controller, achieving current proportionality. Furthermore, the voltage limiter ensures voltage stability within an acceptable range. The specific implementation process is described below.

[0095] First, we introduce the DC microgrid structure and necessary graph theory knowledge:

[0096] like Figure 2As shown, the DC microgrid of this embodiment includes N DC-DC converters, the distributed generation units are connected to the same bus through the DC-DC converters, and the loads are powered by the DC bus.

[0097] N DC-DC converters in a DC microgrid exchange information through a distributed communication network. Consider a communication network containing N converters, defined as G = (V, E, A), where V is a node set, a node is a converter, is the set of all edges, which are communication connections between nodes; is the adjacency matrix, [a ij ] is the weight coefficient, when (v i ,v j )∈E when a ij =1, and all others are 0.

[0098] Define N i is the set of adjacent nodes of the i-th node, D is the degree matrix of the communication network, D in =D out =diag{d i},in The Laplace matrix is defined as L = DA, and when i = j, When i≠j, L ij =-a ij .

[0099] The above graph theory knowledge is the basis for communication between neighboring nodes in distributed controllers. Here, graph theory knowledge refers to the fact that the subsequent distributed controller design and working mode are based on such undirected communication graphs.

[0100] The control method of this embodiment is as follows: Figure 1 As shown in the figure, the DC microgrid is distributed controlled through the voltage-current dual-loop control and the proposed strategy to ensure the proportional distribution of the output current of each converter and ensure that the bus voltage fluctuates within the allowable range.

[0101] The traditional model of DC microgrid is established as follows:

[0102]

[0103] y i =f i (x i )

[0104] where f i (x i )=G i x i , G i =

[0100] . The corresponding matrix represents the output voltage, output current, the difference between the rated voltage and the output voltage, and the difference between the rated current and the output current. The corresponding matrix represents 0, voltage source voltage, current loop proportional coefficient * voltage loop proportional coefficient / inductance, voltage loop proportional coefficient. And

[0105]

[0106]

[0107] And the all-wheel drive method is used to convert it into an all-wheel drive system model, which is expressed as follows:

[0108]

[0109] in, is f i To W i Lie derivative of .

[0110]

[0111] Therefore, the system input, i.e. the final control quantity, can be expressed as

[0112]

[0113] in The control input designed for the subsequent control strategy is set by setting the first item in the brackets as follows

[0114]

[0115] Therefore, the all-wheel drive decoupling prediction model can be finally expressed as follows

[0116]

[0117] Among them I i is the output current on bus i, k represents the discrete time, T s is the system sampling step. According to the all-wheel drive decoupling prediction model, the predicted output current is as follows:

[0118]

[0119] in,

[0120]

[0121] The autoregressive prediction model is used to estimate the predicted value of the control quantity outside the control time domain in the algorithm. The autoregressive prediction model can be summarized as follows:

[0122]

[0123] Among them, N u Represents the control time domain of the control quantity, N y represents the predicted time domain of the output current, v represents the constant coefficient, θ represents the difference between the predicted value and the true value, and κ is the corresponding proportional coefficient.

[0124] According to the control quantity prediction value at the previous moment, the output current prediction value at the current moment, and the unknown control quantity prediction value estimated by the autoregressive model at the current moment, the design cost function is as follows:

[0125]

[0126] Among them, a ij is the communication weight, η i >0 represents a punishment factor.

[0127] make Available

[0128]

[0129] in and

[0130] Finally, the simplified form of the required control quantity is as follows:

[0131]

[0132] Where Λ is a matrix satisfying: and For any i≠j, Ψ is a matrix satisfying: and For any i≠j. s i =[0…0 1].

[0133] The final system input is as follows

[0134]

[0135] Based on the above system input, the voltage limiter is further designed as follows:

[0136]

[0137] in, Represents the voltage deviation value allowed by the system, Represents the bus voltage reference value of the system input.

[0138] Finally, two cases are simulated using the Matlab / Simulink simulation platform to verify the effectiveness of the proposed method. Consider a microgrid circuit consisting of three buck converters, as shown in the schematic diagram. Figure 3 As shown, the parameters of the algorithm are set to α=50,η=8,N y =10, and N u =5, and the rest of the circuit parameters are shown in Table 1:

[0139] Table 1 Circuit parameters

[0140]

[0141] Case 1: Step Change Resistive Load

[0142] In this case study, the following four stages are set to investigate how the control algorithm can achieve the control target under load step changes:

[0143] (1) Phase 1 (0-2s): At t = 0, no public load is connected.

[0144] (2) Phase 2 (2-3s): A common load of 50Ω is connected to the circuit.

[0145] (3) Phase 3 (3-5s): Public load is cut out.

[0146] Case results such as Figure 4 As shown, Figure 4 (a) is the current result diagram, Figure 4 (b) is the voltage result diagram. The experimental results show that the proposed algorithm can guarantee the control performance when the resistive load changes step by step.

[0147] Case 2: Step-Change Constant Power Load

[0148] In this case study, the following four stages are set to investigate how the control algorithm can achieve the control target under constant power load step changes:

[0149] (1) Phase 1 (0-2s): At t = 0, a 500W constant power load is connected.

[0150] (2) Phase 2 (2-3s): 500W constant power load changes to 700W.

[0151] (3) Stage 3 (3-5s): Switch back to the 500W constant power load.

[0152] Case results such as Figure 5 As shown, Figure 5 (a) is the current result diagram, Figure 5(b) is the voltage result diagram. The experimental results show that the proposed algorithm can guarantee the control performance under the condition of constant power load step change.

[0153] Example 2

[0154] The purpose of this embodiment is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of the above method when executing the program.

[0155] Example 3

[0156] The purpose of this embodiment is to provide a computer-readable storage medium.

[0157] A computer-readable storage medium stores a computer program, which, when executed by a processor, performs the steps of the above method.

[0158] Example 4

[0159] The purpose of this embodiment is to provide a distributed full-drive prediction system for a DC microgrid based on an autoregressive model, including:

[0160] The all-wheel drive model prediction module is configured to: perform decoupling analysis on the DC microgrid coupling model to obtain an all-wheel drive decoupling prediction model, and obtain an output current in a prediction time domain based on the all-wheel drive decoupling prediction model;

[0161] The autoregressive model prediction module is configured to: based on the control quantity within the control time domain predicted at the previous moment, use the autoregressive model to further predict the required control quantity prediction value that cannot be directly obtained to obtain the predicted value;

[0162] The predictive control solution module is configured to: design a loss function based on the obtained predicted value and the predicted output current and solve the obtained control variable;

[0163] The voltage limiting module is configured to: design a voltage limiter according to the obtained final control quantity to ensure that the bus voltage reference value fluctuates within an allowable range;

[0164] The microgrid control module is configured to: input the bus voltage reference value corresponding to the above-mentioned voltage limiter into the voltage-current dual-loop controller to make the bus voltage track the bus voltage reference value.

[0165] Example 5

[0166] The purpose of this embodiment is to provide a computer program product containing instructions, which, when running on a computer, enables the computer to execute the methods and functions involved in any of the above embodiments.

[0167] The steps involved in the apparatus of the above embodiment correspond to those of the method embodiment 1. For detailed implementation, please refer to the relevant description of embodiment 1. The term "computer-readable storage medium" should be understood to mean a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium capable of storing, encoding, or carrying an instruction set for execution by a processor and causing the processor to perform any method of the present invention.

[0168] Those skilled in the art will appreciate that the modules or steps of the present invention described above can be implemented using a general-purpose computer device. Alternatively, they can be implemented using program code executable by a computing device, which can then be stored in a storage device and executed by the computing device. Alternatively, they can be fabricated into separate integrated circuit modules, or multiple modules or steps can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0169] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A distributed full-drive prediction method for DC microgrids based on an autoregressive model, characterized by: include: Decoupling analysis of the DC microgrid coupling model is performed to obtain an all-wheel drive decoupling prediction model; The output current in the prediction time domain is obtained based on the all-wheel drive decoupling prediction model. Based on the control quantity in the control time domain predicted at the previous moment, the autoregressive model is used to further predict the required control quantity prediction value that cannot be directly obtained to obtain the predicted value; According to the obtained predicted value, the loss function is designed and the obtained control quantity is solved; Design a voltage limiter based on the final control quantity obtained to ensure that the bus voltage reference value fluctuates within the allowable range; The bus voltage reference value corresponding to the above voltage limiter is input into the voltage-current dual-loop controller to make the bus voltage track the bus voltage reference value.

2. The DC microgrid distributed full drive prediction method based on the autoregressive model according to claim 1 is characterized in that: The DC microgrid coupling model construction process includes: Constructing a first matrix including: a matrix of output voltage, output current, difference between rated voltage and output voltage, and difference between rated current and output current; The second matrix is constructed including: voltage source voltage, current loop proportional coefficient*voltage loop proportional coefficient / inductance, and a matrix of voltage loop proportional coefficient.

3. The distributed full drive prediction method for a DC microgrid based on an autoregressive model according to claim 1, characterized in that: The DC microgrid coupling model is converted into a full-wheel drive decoupling prediction model using a full-wheel drive method.

4. The method for predicting distributed full drive of a DC microgrid based on an autoregressive model according to claim 1, wherein: The autoregressive prediction model is constructed by taking into account: the control time domain of the control variable, the prediction time domain of the output current, the constant coefficient, the difference between the predicted value and the true value, and the proportional coefficient.

5. The DC microgrid distributed full drive prediction method based on the autoregressive model according to claim 1 is characterized in that: The voltage limiter is as follows: where θ i Represents the voltage deviation value allowed by the system, V i * Represents the bus voltage reference value of the system input.

6. The DC microgrid distributed full drive prediction method based on the autoregressive model according to claim 1 is characterized in that: include: Distributed control is performed by setting up controllers at each converter: The controller set at each converter is to build a DC microgrid topology containing multiple converters, set a voltage and current dual-loop control and a controller of the proposed control strategy at each converter, and connect each converter to a distributed communication network.

7. A distributed full-drive prediction system for DC microgrids based on an autoregressive model, characterized by: include: The all-wheel drive model prediction module is configured to: perform decoupling analysis on the DC microgrid coupling model to obtain an all-wheel drive decoupling prediction model, and obtain an output current in a prediction time domain based on the all-wheel drive decoupling prediction model; The autoregressive model prediction module is configured to: based on the control quantity within the control time domain predicted at the previous moment, use the autoregressive model to further predict the required control quantity prediction value that cannot be directly obtained to obtain the predicted value; The predictive control solution module is configured to: design a loss function based on the obtained predicted value and the predicted output current and solve the obtained control variable; The voltage limiting module is configured to: design a voltage limiter according to the obtained final control quantity to ensure that the bus voltage reference value fluctuates within an allowable range; The microgrid control module is configured to: input the bus voltage reference value corresponding to the above-mentioned voltage limiter into the voltage-current dual-loop controller to make the bus voltage track the bus voltage reference value.

8. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the program, the steps of the method described in any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 6 are performed.

Images