Dynamic simulation method of grid-connected converter based on time-varying impedance characteristics of new energy station

Abstract

The present application relates to a kind of based on new energy station time-varying impedance characteristics Grid-connected converter dynamic simulation method, belong to electric power system field.The method is: define the index of measuring new energy station wideband stability, judge whether the impedance amplitude-frequency characteristic of interconnection system full frequency band exists intersection point and the phase margin value at intersection point, according to the margin information provided to assess the small signal stability of system;Establish the model of new energy station grid-connected system, verify the correctness of the model, and obtain the impedance response data under different operating conditions;Based on deep learning off-line, the dynamic simulation mapping model of grid-connected inverter related to the operating point of new energy station is constructed, and the dynamic simulation process of grid-connected inverter control is carried out, if the demand is not met, adjust weight value and retrain until the error requirement is met, finally output the trained model;The impedance characteristics of new energy station are analyzed by impedance analysis method and Nyquist theorem, and the simulation model result is compared and verified.

Description

Technical Field

[0001] This invention belongs to the field of power systems and relates to a dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants. Background Technology

[0002] In actual engineering projects, some units may be disconnected due to random failures or adjustments to the operation plan, resulting in time-varying characteristics of the topology of new energy power stations. This poses a challenge to the aggregation calculation of impedance networks and makes it difficult to analyze the broadband oscillation stability of new energy power stations. Therefore, it is urgent to accurately evaluate and dynamically simulate the time-varying impedance characteristics of new energy power stations.

[0003] With the increasing complexity of power systems and the rapid development of new energy sources, more real-time and accurate methods are needed to assess and maintain the stability of broadband oscillations. Current simulation analysis methods mainly include offline simulation modeling, hardware-in-the-loop (HIL) simulation, and full physical testing. However, these methods struggle to consider the interactions between the power plant's internal and external environments. HIL simulation focuses on impedance scanning of a single unit, and full physical testing can only be applied to a single unit. With the development of communication and information technology, cloud computing and big data analytics can also support the stability analysis of broadband oscillations, enabling more accurate and comprehensive stability assessments. Furthermore, big data analytics can extract potential oscillation characteristics and patterns from massive amounts of data, providing more effective strategies and measures for the suppression and prevention of broadband oscillations. Therefore, this paper proposes a method based on substation converter control to dynamically simulate the time-varying impedance characteristics of new energy power plants.

[0004] Based on the above background, this paper proposes a dynamic simulation method for grid-connected converters of new energy power plants to address the broadband stability problem of grid connection. First, a broadband stability evaluation index for new energy power plants is defined. Then, impedance characteristics and control signal data are obtained through simulation of the grid-connected system of the new energy power plant. The data is then processed to establish a deep learning model that maps impedance response to control signals. Finally, the accuracy of the proposed strategy is verified based on the simulation model, thus achieving dynamic simulation of the grid-connected converter of the new energy power plant. Summary of the Invention

[0005] In view of this, the purpose of this invention is to provide a dynamic simulation method for grid-connected inverters based on the time-varying impedance characteristics of renewable energy power plants. This method enables dynamic simulation of grid-connected inverters in renewable energy power plants, and, based on artificial intelligence, controls the inverters according to different operating conditions, evaluating the wideband stability of grid connection in renewable energy power plants.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants, comprising the following steps:

[0008] S1: Defines an index for measuring the broadband stability of new energy power plants, determines whether there is an intersection point in the impedance amplitude-frequency characteristics of the interconnected system across the entire frequency band, and identifies the phase margin value at the intersection point. Based on the provided margin information, it assesses the small-disturbance stability of the system. System stability depends on the ratio of inverter output impedance to grid impedance.

[0009]

[0010] The system is stable when H(s) satisfies the Nyquist stability criterion, and the stable phase margin is:

[0011] PM(s) = 180° + ∠H(j2πf) int )

[0012] f int The loop gain under different parameter variables and the frequency of the 0dB plane, i.e., the impedance junction frequency;

[0013] S2: Establish a grid-connected system model for a new energy power plant. The system model consists of direct-drive wind turbines, a grid-connected inverter, and a grid module. The direct-drive wind turbine is equivalent to a current source, connected to the grid-connected inverter via a DC bus capacitor. The output is connected to a 690V / 35kV transformer, and after step-up, the voltage is collected and then connected to a 35kV / 220kV transformer via a collector line. Finally, it is connected to the grid module via a submarine cable transmission line to achieve grid connection. The grid-connected inverter adopts a follower-type control. The system model data includes: i w1 This refers to the three-phase current on the grid-connected side of the direct-drive wind turbine generator; v wc i wc The three-phase voltage and three-phase current on the filter capacitor side; v w2 i w2 For the grid-connected converter, the three-phase voltage and three-phase current on the grid-connected side; I w Represents the equivalent current source, Uw is the DC bus voltage, C w For DC side capacitor; m abc It is a three-phase modulated signal, i.e., θ p T is the synchronization phase angle output by the PLL after sampling the grid-connected voltage. abc / dq T dq / abc These are the abc-dq transformation matrix and the dq-abc transformation matrix, respectively, used to generate the electrical quantities in the dq coordinate system of the input controller and the electrical quantities in the abc coordinate system of the controller output to the electrical model; s is the Laplace operator; L f1 The inductor on the converter side of the LCL filter has an impedance Zf1 = sLf1; Cf and Rf2 are the filter capacitors and their series resistance; L f2The grid-side filter inductance and equivalent resistance are shown; values ​​marked with * are rated values. The operation of the new energy power station is simulated in Simulink to verify the correctness of the established model and obtain impedance response data under different operating conditions.

[0014] S3: For the system model data provided in S2, part is used as the training set and the other part as the test set. Based on deep learning, an offline dynamic simulation mapping model of the grid-connected inverter of the new energy power station is constructed. The impedance response data of the new energy power station, including impedance amplitude, phase and frequency, is used as input, and the control signal v of the grid-connected inverter of the new energy power station is used as input. wd v wq As output, the grid-connected inverter control dynamic simulation process, if the requirements are not met, adjust the weight values ​​and retrain until the error requirements are met, and finally output the trained model;

[0015] S4: Based on the model trained as described in S3, and according to the online operating conditions of the grid-connected system of the new energy power station, the known impedance response data of the new energy power station are obtained. The impedance analysis method and Nyquist theorem are used to conduct stability analysis on the impedance characteristics of the new energy power station, and the results are compared and verified with the simulation model.

[0016] Furthermore, the dynamic simulation of grid-connected inverter control specifically includes:

[0017] First, the output impedance response data of the grid-connected system is input. The input data is normalized, and an appropriate number of implicit layers and initial weight values ​​are set. The control parameters of the grid-connected inverter of the new energy power station are output and compared with the actual output values. The mean square error (MSE) is used to evaluate the prediction effect, as shown in the following formula:

[0018]

[0019] Furthermore, the simulation model results comparison and verification are specifically as follows: Taking the offshore wind farm grid-connected system as the research object, 80 wind turbine units with a rated total capacity of 400MW are used to establish a simulation model. Given the known impedance characteristics and frequency sweep results of the new energy power station, the amplitude, phase, and frequency of the positive-sequence impedance and negative-sequence impedance characteristics are extracted respectively. By changing the parameters of the grid-connected converter, impedance characteristic data under different operating conditions are obtained. A deep learning model is established, taking the impedance response data of the new energy power station, including impedance amplitude, phase, and frequency, as input, and the control signal of the grid-connected inverter of the new energy power station as output. By adjusting the network hyperparameters, the error on the test set is minimized based on the loss function, thereby verifying the accuracy of the model.

[0020] Furthermore, one portion is 80% used as the training set, and the other portion is 20% used as the test set.

[0021] The beneficial effects of this invention are as follows:

[0022] This invention proposes a dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of renewable energy power plants, utilizing an intelligent optimization method for grid-connected converter control signals, based on impedance analysis stability. The main advantages are twofold:

[0023] (1) This invention analyzes the time-varying and offline nature of the impedance analysis method used in broadband stability analysis. The parameters are ideal, but it is difficult to reflect the interaction between new energy power plants and the power grid. This proves that the online dynamic simulation method of grid-connected inverters is of great significance for the stable operation of grid-connected systems of new energy power plants.

[0024] (2) This invention proves the accuracy of the output impedance characteristics of the grid-connected system of new energy power stations based on artificial intelligence simulation. Compared with single physical simulation, the time-varying impedance characteristics of new energy power stations using data-driven and software-defined methods can better reflect the characteristics of the power station than single physical simulation.

[0025] Other advantages, objectives, and features of the invention will be set forth in part in the description which follows, and in part will be apparent to those skilled in the art from the following examination, or may be learned from practice of the invention. The objectives and other advantages of the invention can be realized and obtained through the following description. Attached Figure Description

[0026] To make the objectives, technical solutions, and advantages of the present invention clearer, the preferred embodiments of the present invention will be described in detail below with reference to the accompanying drawings, wherein:

[0027] Figure 1 This is the main circuit topology for the grid connection system of new energy power plants;

[0028] Figure 2 Structure of grid-connected converter controller for new energy power plants;

[0029] Figure 3 A flowchart for dynamic simulation control of grid-connected inverters in new energy power plants;

[0030] Figure 4 The impedance characteristics and frequency sweep results of the known new energy power stations;

[0031] Figure 5 The training effect of the dynamic simulation model for grid-connected converters. Detailed Implementation

[0032] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Unless otherwise specified, the following embodiments and features can be combined with each other.

[0033] The accompanying drawings are for illustrative purposes only and are schematic diagrams, not actual pictures. They should not be construed as limiting the invention. To better illustrate the embodiments of the invention, some parts in the drawings may be omitted, enlarged, or reduced, and do not represent the actual product dimensions. It is understandable to those skilled in the art that some well-known structures and their descriptions may be omitted in the drawings.

[0034] In the accompanying drawings of the embodiments of the present invention, the same or similar reference numerals correspond to the same or similar components. In the description of the present invention, it should be understood that if terms such as "upper," "lower," "left," "right," "front," and "rear" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting the present invention. For those skilled in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0035] This invention relates to a data-model-driven broadband stability online control method for renewable energy power plants. The method includes the following steps:

[0036] S1: Defines an index for measuring the broadband stability of new energy power plants, determines whether there is an intersection point in the impedance amplitude-frequency characteristics of the interconnected system across the entire frequency band, and identifies the phase margin value at the intersection point. Based on the provided margin information, it assesses the small-disturbance stability of the system. System stability depends on the ratio of inverter output impedance to grid impedance.

[0037]

[0038] The system is stable when H(s) satisfies the Nyquist stability criterion, and the stable phase margin is:

[0039] PM(s) = 180° + ∠H(j2πf) int )

[0040] f intThe loop gain and the frequency of the 0dB plane under different parameter variables are the impedance junction frequency.

[0041] S2: Establish a grid connection system model for new energy power plants, such as... Figure 1 As shown, i w1 This refers to the three-phase current on the grid-connected side of the direct-drive wind turbine. wc i wc These represent the three-phase voltage and three-phase current on the filter capacitor side. w2 i w2 This refers to the three-phase voltage and three-phase current on the grid-connected side of the grid-connected converter. w Represents the equivalent current source, Uw is the DC bus voltage, C w This is the DC-side capacitor. abc This is a three-phase modulated signal, i.e., θ. p T is the synchronization phase angle output by the PLL after sampling the grid-connected voltage. abc / dq T dq / abc These are the abc-dq transformation matrices and the dq-abc transformation matrices, respectively, used to generate the electrical quantities in the dq coordinate system of the input controller and the electrical quantities in the abc coordinate system of the controller output to the electrical model. s is the Laplace operator. L f1 The inductor on the converter side of the LCL filter has an impedance Zf1 = sLf1. Cf and Rf2 are the filter capacitors and their series resistance. f2 This represents the grid-side filter inductance and equivalent resistance. Values ​​marked with an asterisk (*) are rated values. The operation of the renewable energy power station is simulated in Simulink to verify the correctness of the model and obtain impedance response data under different operating conditions. Figure 2 The structure of the grid-connected converter controller for new energy power plants.

[0042] S3: Based on the data provided in S2, 80% is used as the training set and 20% as the test set. A dynamic simulation mapping model of the operating point-dependent grid-connected inverter for new energy power plants is constructed offline using deep learning. The impedance response data of the new energy power plants, including impedance amplitude, phase, and frequency, is used as input, along with the control signal v of the grid-connected inverter. wd v wq As output, such as Figure 3 The diagram shows the dynamic simulation flowchart for grid-connected inverter control. First, the output impedance response data of the grid-connected system is input. The input data is normalized, and appropriate implicit layers and initial weight values ​​are set. The control parameters of the grid-connected inverter for the new energy power plant are output and compared with the actual output values. The mean square error (MSE) is used to evaluate the prediction effect, as shown in the following formula:

[0043]

[0044] If the requirements are not met, the weight values ​​are adjusted and the training is repeated until the error requirements are met, and finally the trained model is output.

[0045] S4: Based on the deep learning model trained in S3, and according to the online operating conditions of the grid-connected system of new energy power plants, the known impedance response data of new energy power plants are obtained. The impedance characteristics of the new energy power plants are analyzed for stability using impedance analysis and Nyquist theorem, and the results are compared and verified with the simulation model.

[0046] Taking an offshore wind farm grid-connected system as the research object, a simulation model is established using 80 wind turbine units with a total rated capacity of 400MW, such as... Figure 4 As shown, the known impedance characteristics and frequency sweep results of new energy power plants are presented. The amplitude, phase, and frequency of the positive-sequence and negative-sequence impedance characteristics are extracted respectively. By changing the parameters of the grid-connected inverter, impedance characteristic data under different operating conditions are obtained. A deep learning model is established, using the impedance response data of the new energy power plant, including impedance amplitude, phase, and frequency, as input, and the control signal of the grid-connected inverter as output. By adjusting the network hyperparameters and minimizing the error on the test set based on the loss function, the accuracy of the model is verified. Figure 5 The results shown are the training effects of the dynamic simulation model of the grid-connected converter, which are highly accurate and meet the requirements.

[0047] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants, characterized in that: The method includes the following steps: S1: Defines an index to measure the broadband stability of new energy power plants, determines whether there is a boundary point in the impedance amplitude-frequency characteristics of the interconnected system across the entire frequency band, and the phase margin value at the boundary point. Based on the provided margin information, it evaluates the small-disturbance stability of the system. The system stability depends on the ratio of the inverter output impedance to the grid impedance. when H ( s The system is stable when the Nyquist stability criterion is satisfied, and the stable phase margin is: f int The loop gain under different parameter variables and the frequency of the 0dB plane, i.e., the impedance junction frequency; S2: Establish a grid-connected system model for a new energy power plant. The system model consists of direct-drive wind turbines, a grid-connected inverter, and a grid module. The direct-drive wind turbine is equivalent to a current source, connected to the grid-connected inverter via a DC bus capacitor. The output is connected to a 690V / 35kV transformer, and after step-up, the voltage is collected and then connected to a 35kV / 220kV transformer via a collector line. Finally, it is connected to the grid module via a submarine cable transmission line to achieve grid connection. The grid-connected inverter adopts a follower-type control. The system model data includes: i w1 This refers to the three-phase current on the grid-connected side of the direct-drive wind turbine generator; v wc i wc The three-phase voltage and three-phase current on the filter capacitor side; v w2 i w2 For the grid-connected converter, the three-phase voltage and three-phase current on the grid-connected side; I w Represents the equivalent current source, Uw is the DC bus voltage, C w For DC side capacitors; m abc It is a three-phase modulated signal; θ p T is the synchronization phase angle output by the PLL after sampling the grid-connected voltage. abc / dq T dq / abc These are the abc-dq transformation matrix and the dq-abc transformation matrix, respectively, used to generate the electrical quantities in the dq coordinate system of the input controller and the electrical quantities in the abc coordinate system of the controller output to the electrical model; s is the Laplace operator; L f1 The inductor on the converter side of the LCL filter has an impedance of C f R f2 These are the filter capacitor and its series resistor, respectively; L f2 The grid-side filter inductor and equivalent resistance are used; the operation of the new energy power station is simulated in Simulink to verify the correctness of the established system model and obtain impedance response data under different operating conditions. S3: For the system model data provided in S2, part is used as the training set and the other part as the test set. Based on deep learning, an offline dynamic simulation mapping model of the grid-connected inverter of the new energy power station is constructed. The impedance response data of the new energy power station, including impedance amplitude, phase and frequency, is used as input, and the control signal v of the grid-connected inverter of the new energy power station is used as input. wd v wq As output, the grid-connected inverter control dynamic simulation process uses the root mean square error to evaluate the error between the predicted result and the actual output value. When the error does not meet the error requirements, the weight values ​​of the dynamic simulation mapping model are adjusted and retrained until the error meets the error requirements, and the trained model is output. S4: Based on the model trained as described in S3, and according to the online operating conditions of the grid-connected system of the new energy power station, the known impedance response data of the new energy power station are obtained. The impedance analysis method and Nyquist theorem are used to conduct stability analysis on the impedance characteristics of the new energy power station, and the results are compared and verified with the simulation model.

2. The dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants according to claim 1, characterized in that: The specific dynamic simulation of grid-connected inverter control is as follows: First, the output impedance response data of the grid-connected system is input. The input data is normalized, and an appropriate number of implicit layers and initial weight values ​​are set. The control parameters of the grid-connected inverter of the new energy power station are output and compared with the actual output values. The root mean square error (RMSE) is used to evaluate the prediction effect, as shown in the following formula: 。 3. The dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants according to claim 2, characterized in that: The simulation model results comparison and verification are specifically as follows: Taking the offshore wind farm grid-connected system as the research object, 80 wind turbine units with a rated total capacity of 400MW are used to establish a simulation model. Given the known impedance characteristics and frequency sweep results of the new energy power station, the amplitude, phase, and frequency of the positive-sequence impedance and negative-sequence impedance characteristics are extracted respectively. By changing the parameters of the grid-connected converter, impedance characteristic data under different operating conditions are obtained. A deep learning model is established, and the impedance response data of the new energy power station, including impedance amplitude, phase, and frequency, are used as input, and the control signal of the grid-connected inverter of the new energy power station is used as output. By adjusting the network hyperparameters, the error on the test set is minimized based on the loss function, thereby verifying the accuracy of the deep learning model.

4. The dynamic simulation method for grid-connected converters based on the time-varying impedance characteristics of new energy power plants according to claim 1, characterized in that: One portion is 80% used as the training set, and the other portion is 20% used as the test set.

Images