A method, device and storage medium for improving synchronous oscillation of a new energy synchronous machine

Abstract

The present invention relates to a method, device and storage medium for improving synchronous oscillation of a new energy synchronous machine, and relates to the field of three-phase inverter control. The present invention models the synchronous machine to obtain a dynamic model of the synchronous machine; obtains the virtual rotor angular velocity of the synchronous machine for each control step based on the dynamic model of the synchronous machine; and integrates the virtual rotor angular velocity to obtain the voltage phase of the three-phase inverter relative to the power grid; obtains the predicted voltage amplitude based on the voltage phase of the three-phase inverter relative to the power grid, the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the power grid through a pre-trained voltage amplitude control network; selects the predicted voltage amplitude as the target voltage amplitude according to the Bellman equation that minimizes the evaluation function, and generates a signal to drive the three-phase inverter based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the power grid.

Description

Technical Field

[0001] The present invention relates to the field of three-phase inverter control, and in particular to a method, a device and a storage medium for improving synchronous oscillation of a new energy synchronous machine. Background Art

[0002] With the development of new energy technologies, a large amount of electric energy generated by new energy is directly or indirectly connected to the grid through inverters after being stored in energy storage. When the inverter is negligible compared with the scale of the grid, there will be no oscillation problem. However, when the inverter is connected to a weak grid or microgrid, it will cause synchronous oscillation problems. The essence of the sub-synchronous oscillation problem of photovoltaic stations lies in the fact that the highly power electronic grid-connected interface weakens the damping of the original system. For inductive high-voltage grids, active power is proportional to the voltage phase of the three-phase inverter relative to the grid, and reactive power is proportional to the inverter voltage. However, in low-voltage grids such as weak grids or microgrids, most of the reactive power of the resistive or semi-resistive grid depends on the voltage phase and voltage amplitude of the three-phase inverter relative to the grid. In addition, in the power system, the inverter may face some uncertain factors, such as line impedance changes or nonlinear behavior of electrical components, which will change the reactive power relationship. The use of linear controllers to adjust the inverter voltage based on reactive power will cause the damping characteristics to be unmatched, causing oscillation. Summary of the invention

[0003] In order to solve the above technical problem or at least partially solve the above technical problem, the present invention provides a method, a device and a storage medium for improving synchronous oscillation of a new energy synchronous machine.

[0004] In a first aspect, the present invention provides a method for improving synchronous oscillation of a new energy synchronous machine, which is applied to an inverter grid-connected circuit, comprising:

[0005] The synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine; based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine in each control step is obtained ; and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid ;

[0006] The voltage amplitude control network is pre-trained based on the voltage phase of the three-phase inverter relative to the grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the power grid are used to obtain the predicted voltage amplitude; wherein, the voltage phase of the three-phase inverter relative to the power grid is obtained through the pre-trained evaluation network. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid and the predicted voltage amplitude to predict the value of the evaluation function after the state transfer according to the predicted voltage amplitude;

[0007] According to the Bellman equation that minimizes the evaluation function, the predicted voltage amplitude is selected as the target voltage amplitude. The PWM generator is based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the grid. Generate three-phase PWM signals to drive the three-phase inverter.

[0008] Furthermore, the synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine as follows:

[0009] ;

[0010] in, is the output power of the synchronous machine, is the input power of the synchronous machine, is the virtual rotor inertia moment of the synchronous machine, is the virtual rotor angular velocity of the synchronous machine, is the damping coefficient, is the angular velocity of the grid when the inverter is connected to the grid.

[0011] Furthermore, obtaining the virtual rotor angular velocity of the synchronous machine for each control step based on the dynamic model of the synchronous machine includes: presetting the input power of the synchronous machine and the damping coefficient , through the output voltage of the three-phase inverter and output current Calculate the output power of synchronous machine , collect the angular velocity of the grid when the inverter is connected to the grid ; Based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine is obtained for each control step ; and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid .

[0012] Furthermore, the active power transmitted by the three-phase inverter to the grid is and reactive power for:

[0013] ;

[0014] ;

[0015] in, is the peak value of the three-phase inverter output voltage, is the peak value of the grid voltage.

[0016] Furthermore, the inverter grid-connected circuit includes a DC output of a new energy source, a DC output of the new energy source is connected in parallel with an energy storage capacitor, the DC output of the new energy source is connected to a three-phase inverter, the output of the three-phase inverter is connected to a second-order LC filter, the three-phase inverter obtains power from a DC power source and injects electrical energy into the grid in the form of AC; in the second-order LC filter, the inductive reactance is and the capacitive reactance is ; Through the reactance The inductance and resistance are The resistance modeling transmission line; then the equivalent reactance of each phase of the three-phase inverter is , the equivalent impedance of each phase is .

[0017] Furthermore, the voltage amplitude control network and the evaluation network are fully connected networks;

[0018] The voltage amplitude control network comprises 6 input nodes, which receive the voltage phase of the three-phase inverter relative to the power grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the grid; set the number of hidden layers and 1 output node; the output node outputs the predicted voltage amplitude;

[0019] The evaluation network includes 7 input nodes, which receive the voltage phase of the three-phase inverter relative to the grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid, and the predicted voltage amplitude; a set number of hidden layers and 1 output node, the output node outputs the value of the evaluation function after state transfer according to the predicted voltage amplitude.

[0020] Furthermore, the loss function for training the evaluation network is the mean square error between the value of the evaluation function after state transfer according to the predicted voltage amplitude and the true evaluation function value. The evaluation network is trained by a gradient descent algorithm with the goal of minimizing the loss function. The true evaluation function value is the accumulation of the weighted sum of the active power error of the three-phase inverter after each decision step, the reactive power error of the three-phase inverter, and the frequency error between the three-phase inverter and the power grid.

[0021] Furthermore, the voltage amplitude control network aims to predict the voltage amplitude, with the goal of minimizing the Bellman equation of the evaluation function within a set time period, and adopts a back propagation algorithm to update the weights of the voltage amplitude control network.

[0022] In a second aspect, the present invention provides a device for improving synchronous oscillation of a new energy synchronous machine, comprising: a DC output of a new energy, a DC output of the new energy connected in parallel with an energy storage capacitor, the DC output of the new energy connected to a three-phase inverter, the three-phase inverter obtains power from a DC power supply and injects electrical energy into a power grid in the form of AC; the three-phase inverter is electrically connected to a PWM generator, the PWM generator is connected to a controller, the controller includes at least one processing unit, the processing unit, a storage unit and a collection unit are interconnected through a bus unit, the storage unit stores a computer program and data collected by the collection unit, and when the computer program is executed by the processing unit, the method for improving synchronous oscillation of a new energy synchronous machine is implemented.

[0023] In a third aspect, the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method for improving synchronous oscillation of a new energy synchronous machine is implemented.

[0024] The above technical solution provided by the embodiment of the present invention has the following advantages compared with the prior art:

[0025] The present application models the synchronous machine simulated by the three-phase inverter to obtain a dynamic model of the synchronous machine; based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine in each control step is obtained, and the virtual rotor angular velocity is integrated to obtain the voltage phase of the three-phase inverter relative to the power grid; through a pre-trained voltage amplitude control network, a predicted voltage amplitude is obtained based on the voltage phase of the three-phase inverter relative to the power grid, the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the power grid; through a pre-trained voltage amplitude control network, a predicted voltage amplitude is obtained based on the voltage phase of the three-phase inverter relative to the power grid, the active power of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the power grid; The trained evaluation network predicts the value of the evaluation function after state transfer according to the predicted voltage amplitude based on the voltage phase of the three-phase inverter relative to the power grid, the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid, and the predicted voltage amplitude; the predicted voltage amplitude is selected as the target voltage amplitude according to the Bellman equation that minimizes the evaluation function, and the PWM generator generates a three-phase PWM signal to drive the three-phase inverter based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the power grid. For the frequency control part, the present application still solves based on active power, and for the voltage amplitude part, the present application introduces a pre-trained voltage amplitude control network and a pre-trained evaluation network, and performs nonlinear modeling through heterogeneous models of decision-making and evaluation, which can effectively model the nonlinear relationship between the voltage phase and voltage amplitude of the three-phase inverter relative to the power grid, thereby performing adaptive damping reactive power control based on the predicted voltage amplitude to better achieve steady state. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention.

[0027] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0028] Figure 1 A flow chart of a method for improving synchronous oscillation of a new energy synchronous machine provided by an embodiment of the present invention;

[0029] Figure 2 A schematic diagram of an inverter grid-connected circuit provided by an embodiment of the present invention;

[0030] Figure 3 An architecture diagram of a voltage amplitude control network and an evaluation network provided in an embodiment of the present invention;

[0031] Figure 4 A schematic diagram of a device for improving synchronous oscillation of a new energy synchronous machine provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0032] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. 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.

[0033] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0034] Example 1

[0035] like Figure 1 As shown, the technology of the present invention realizes a method for improving synchronous oscillation of a new energy synchronous machine, which is applied to Figure 2 In the new energy inverter grid-connected circuit shown in the figure. In the figure, the inverter grid-connected circuit includes the DC output Vs of the new energy, the DC output of the new energy is connected in parallel with the energy storage capacitor, the DC output of the new energy is connected to the three-phase inverter, and the three-phase inverter formed by the insulated gate bipolar transistors Q1-Q6 obtains power from the DC power supply and injects electrical energy into the grid in the form of AC. The switching characteristics of the inverter will generate high-frequency harmonics, and a low-pass filter is required to eliminate the high-frequency harmonics. The low-pass filter uses a second-order LC filter. In the second-order LC filter, the inductive reactance is and the capacitive reactance is . Through the reactance The inductance and resistance are The equivalent reactance of each phase of the three-phase inverter including the second-order LC filter and the transmission line is , the equivalent impedance of each phase is .

[0036] The synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine:

[0037] ;

[0038] in, is the output power of the synchronous machine, is the input power of the synchronous machine, is the virtual rotor inertia moment of the synchronous machine, is the virtual rotor angular velocity of the synchronous machine, is the damping coefficient, is the angular velocity difference between the inverter and the grid when connected to the grid, , is the angular velocity of the grid when the inverter is connected to the grid. Among them, the output voltage of the three-phase inverter and output current Calculate the output power of synchronous machine .

[0039] Obtain the virtual rotor angular velocity of the synchronous machine for each control step based on the dynamic model of the synchronous machine , and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid Includes: Preset synchronous machine input power and the damping coefficient , through the output voltage of the three-phase inverter and output current Calculate the output power of synchronous machine , collect the angular velocity of the grid when the inverter is connected to the grid ; Based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine is obtained for each control step ; and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid .

[0040] The active power transmitted by the three-phase inverter to the grid is and reactive power for:

[0041] ;

[0042] ;

[0043] in, is the peak value of the three-phase inverter output voltage, is the peak value of the grid voltage, is the phase angle between the three-phase inverter voltage and the grid voltage.

[0044] like Figure 3 As shown, the voltage amplitude control network is pre-trained based on the voltage phase of the three-phase inverter relative to the grid. , Active power of three-phase inverter , reactive power of three-phase inverter , the active power error of the three-phase inverter , reactive power error of three-phase inverter The frequency error between the three-phase inverter and the grid The predicted voltage amplitude is obtained; wherein, the voltage phase of the three-phase inverter relative to the grid is calculated based on the pre-trained evaluation network. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid and the predicted voltage amplitude are used to predict the value of the evaluation function after state transfer according to the predicted voltage amplitude.

[0045] In a specific implementation process, the voltage amplitude control network and the evaluation network are fully connected networks;

[0046] The voltage amplitude control network comprises 6 input nodes, which receive the voltage phase of the three-phase inverter relative to the power grid. , Active power of three-phase inverter , reactive power of three-phase inverter , the active power error of the three-phase inverter , reactive power error of three-phase inverter The frequency error between the three-phase inverter and the grid ; Set the number of hidden layers and 1 output node; the output node outputs the predicted voltage amplitude;

[0047] The evaluation network includes 7 input nodes, which receive the voltage phase of the three-phase inverter relative to the grid. , Active power of three-phase inverter , reactive power of three-phase inverter , the active power error of the three-phase inverter , reactive power error of three-phase inverter The frequency error between the three-phase inverter and the grid and predicted voltage amplitude; set a number of hidden layers and 1 output node, the output node outputs the value of the evaluation function after state transfer according to the predicted voltage amplitude.

[0048] The loss function for training the evaluation network is the mean square error between the value of the evaluation function after state transfer according to the predicted voltage amplitude and the real evaluation function value. The evaluation network is trained by a gradient descent algorithm with the goal of minimizing the loss function. The real evaluation function value is the accumulation of the weighted sum of the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, and the frequency error between the three-phase inverter and the power grid after each decision step.

[0049] The goal of the voltage amplitude control network is to predict the voltage amplitude, with the goal of minimizing the evaluation function within a set time period, and the back propagation algorithm is used to update the weights of the voltage amplitude control network.

[0050] Example 2

[0051] See also Figure 2 and Figure 4 As shown, an embodiment of the present invention provides a device for improving synchronous oscillation of a new energy synchronous machine, comprising: a DC output of a new energy source, a DC output of the new energy source connected in parallel with an energy storage capacitor, a DC output of the new energy source connected to a three-phase inverter, the three-phase inverter obtains power from a DC power supply and injects electric energy into the power grid in the form of AC; the three-phase inverter is electrically connected to a PWM generator, the PWM generator is connected to a controller, the controller includes at least one processing unit, the processing unit and the storage unit are interconnected through a bus unit, the storage unit is a computer-readable storage medium, and can be used to store software programs, computer executable programs and modules, such as the software programs, computer executable programs and modules corresponding to the method for improving synchronous oscillation of a new energy synchronous machine in an embodiment of the present invention. The processing unit implements the above-mentioned method for improving synchronous oscillation of a new energy synchronous machine by running the software programs, computer executable programs and modules stored in the storage unit, including:

[0052] The synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine; based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine in each control step is obtained , and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid ;

[0053] The voltage amplitude control network is pre-trained based on the voltage phase of the three-phase inverter relative to the grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the grid to obtain the predicted voltage amplitude; the voltage phase of the three-phase inverter relative to the grid is obtained through the pre-trained evaluation network , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid and the predicted voltage amplitude to predict the value of the evaluation function after the state transfer according to the predicted voltage amplitude;

[0054] According to the Bellman equation that minimizes the evaluation function, the predicted voltage amplitude is selected as the target voltage amplitude. The PWM generator is based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the grid. Generate three-phase PWM signals to drive the three-phase inverter.

[0055] Of course, the computer program stored in the storage unit of the device for improving synchronous oscillation of a new energy synchronous machine provided in an embodiment of the present invention is not limited to the method operations described above, and can also execute related operations in the method for improving synchronous oscillation of a new energy synchronous machine provided in any embodiment of the present invention.

[0056] Example 3

[0057] An embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program. When the computer program is executed, the method for improving synchronous oscillation of a new energy synchronous machine is implemented, including:

[0058] The synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine; based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine in each control step is obtained , and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid ;

[0059] The voltage amplitude control network is pre-trained based on the voltage phase of the three-phase inverter relative to the grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the grid to obtain the predicted voltage amplitude; the voltage phase of the three-phase inverter relative to the grid is obtained through the pre-trained evaluation network , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid and the predicted voltage amplitude to predict the value of the evaluation function after the state transfer according to the predicted voltage amplitude;

[0060] According to the Bellman equation that minimizes the evaluation function, the predicted voltage amplitude is selected as the target voltage amplitude. The PWM generator is based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the grid. Generate three-phase PWM signals to drive the three-phase inverter.

[0061] A computer-readable storage medium provided in an embodiment of the present invention stores a computer program which is not limited to the method operations described above, but can also execute related operations in a method for improving synchronous oscillation of a new energy synchronous machine provided in any embodiment of the present invention.

[0062] In the embodiments provided by the present invention, it should be understood that the disclosed structures and methods can be implemented in other ways. For example, the structural embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, structures or units, which can be electrical, mechanical or other forms.

[0063] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0064] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0065] The foregoing is merely a specific embodiment of the present invention, which enables those skilled in the art to understand or implement the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but rather to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for improving synchronous oscillation of a new energy synchronous machine, applied to an inverter grid-connected circuit, characterized in that: include: The synchronous machine simulated by the three-phase inverter is modeled to obtain the dynamic model of the synchronous machine: The dynamic model of the synchronous machine simulated by the three-phase inverter is modeled as follows: ； in, is the output power of the synchronous machine, is the input power of the synchronous machine, is the virtual rotor inertia moment of the synchronous machine, is the virtual rotor angular velocity of the synchronous machine, is the damping coefficient, is the angular velocity of the grid when the inverter is connected to the grid; Obtain the virtual rotor angular velocity of the synchronous machine for each control step based on the dynamic model of the synchronous machine , and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid ; Among them, the input power of the preset synchronous machine and the damping coefficient , through the output voltage of the three-phase inverter and output current Calculate the output power of synchronous machine , collect the angular velocity of the grid when the inverter is connected to the grid ; Based on the dynamic model of the synchronous machine, the virtual rotor angular velocity of the synchronous machine is obtained for each control step ; and the virtual rotor angular velocity Integrate to get the voltage phase of the three-phase inverter relative to the grid ; The voltage amplitude control network is pre-trained based on the voltage phase of the three-phase inverter relative to the grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter and the frequency error between the three-phase inverter and the grid are used to obtain the predicted voltage amplitude; the voltage phase of the three-phase inverter relative to the grid is obtained through the pre-trained evaluation network , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid and the predicted voltage amplitude to predict the value of the evaluation function after the state transfer according to the predicted voltage amplitude; The voltage amplitude control network and the evaluation network are fully connected networks; the voltage amplitude control network includes 6 input nodes, which receive the voltage phase of the three-phase inverter relative to the power grid. , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid; a set number of hidden layers and 1 output node; the output node outputs the predicted voltage amplitude; the evaluation network includes 7 input nodes, and the 7 input nodes receive the voltage phase of the three-phase inverter relative to the power grid , the active power of the three-phase inverter, the reactive power of the three-phase inverter, the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, the frequency error between the three-phase inverter and the power grid, and the predicted voltage amplitude; a set number of hidden layers and 1 output node, the output node outputs the value of the evaluation function after state transfer according to the predicted voltage amplitude; According to the Bellman equation that minimizes the evaluation function, the predicted voltage amplitude is selected as the target voltage amplitude. The PWM generator is based on the target voltage amplitude and the voltage phase of the three-phase inverter relative to the grid. Generate three-phase PWM signals to drive the three-phase inverter.

2. The method for improving synchronous oscillation of a new energy synchronous machine according to claim 1, characterized in that: Active power delivered by three-phase inverter to the grid and reactive power for: ； ； in, is the peak value of the three-phase inverter output voltage, is the peak value of the grid voltage.

3. The method for improving synchronous oscillation of a new energy synchronous machine according to claim 1, characterized in that: The inverter grid-connected circuit includes a DC output of a new energy source, a DC output of a new energy source connected in parallel with an energy storage capacitor, a DC output of a new energy source connected to a three-phase inverter, an output of a three-phase inverter connected to a second-order LC filter, and a three-phase inverter that obtains power from a DC power source and injects electrical energy into the grid in the form of AC. In the second-order LC filter, the inductive reactance is and the capacitive reactance is ; Through the reactance The inductance and resistance are The resistance modeling transmission line; then the equivalent reactance of each phase of the three-phase inverter is , the equivalent impedance of each phase is .

4. The method for improving synchronous oscillation of a new energy synchronous machine according to claim 1, characterized in that: The loss function for training the evaluation network is the mean square error between the value of the evaluation function after state transfer according to the predicted voltage amplitude and the true evaluation function value. The evaluation network is trained by a gradient descent algorithm with the goal of minimizing the loss function. The true evaluation function value is the accumulation of the weighted sum of the active power error of the three-phase inverter, the reactive power error of the three-phase inverter, and the frequency error between the three-phase inverter and the power grid after each decision step.

5. The method for improving synchronous oscillation of a new energy synchronous machine according to claim 1, characterized in that: The goal of the voltage amplitude control network is to predict the voltage amplitude, with the goal of minimizing the Bellman equation of the evaluation function within a set time period, and using a back propagation algorithm to update the weights of the voltage amplitude control network.

6. A device for improving synchronous oscillation of a new energy synchronous machine, characterized in that: include: A DC output of new energy, a DC output of new energy connected in parallel with an energy storage capacitor, a DC output of new energy connected to a three-phase inverter, the three-phase inverter obtains power from a DC power supply and injects electrical energy into the power grid in the form of AC; the three-phase inverter is electrically connected to a PWM generator, the PWM generator is connected to a controller, the controller includes at least one processing unit, the processing unit, a storage unit and a collection unit are interconnected through a bus unit, the storage unit stores a computer program and data collected by the collection unit, and when the computer program is executed by the processing unit, the method for improving synchronous oscillation of a new energy synchronous machine as described in any one of claims 1-5 is implemented.

7. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for improving synchronous oscillation of a new energy synchronous machine as described in any one of claims 1 to 5 is implemented.

Images