Impedance identification method and system based on physical information neural network

Abstract

The invention discloses an impedance identification method and system based on a physical information neural network. The method comprises the following steps: firstly, obtaining working point information and original impedance data of a preprocessed MMC; then selecting corresponding control physical information to train a neural network according to the purpose of impedance identification based on the working point information to obtain an initial physical information neural network model, and outputting training control physical information; training an initial physical information neural network model based on working point information and the training control physical information, and outputting training impedance data; comparing the training impedance data with original impedance data to obtain a training error, and adjusting the initial physical information neural network model according to the error to obtain an impedance identification model; and finally, impedance identification is completed by using the impedance identification model. According to the invention, the training accuracy of the neural network impedance modeling method can be improved, and the error of impedance identification is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and in particular to an impedance identification method, system, equipment and medium based on physical information neural network. Background Art

[0002] With the rapid development of new power systems, power electronic equipment is widely used in power systems. At the same time, oscillation events dominated by power electronic equipment occur frequently, which seriously affects the safe and stable operation of power systems. The stability analysis method based on impedance transfer function can be used to solve the oscillation problem of power systems with a high proportion of power electronic equipment. Among them, the impedance modeling method is an important method for stability analysis of power grids and converters. It has the characteristics of intuitive physical meaning and simple analysis.

[0003] At present, there are a few existing technologies that carry out impedance identification of power electronic equipment based on the generalization ability of artificial neural networks. Among them, powerful nonlinear processing tools such as neural networks can be used to train according to existing data to obtain neural network impedance identification models under multiple working points. However, these traditional neural network impedance identification methods produce large errors at the impedance mutation point for impedance identification of MMC (modular multilevel converter) with many resonant peaks, resulting in low accuracy of MMC impedance identification. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide an impedance identification method, system, device and medium based on physical information neural network, which can enhance the connection between input working point information and output impedance information and improve the accuracy of impedance identification.

[0005] In order to solve the above technical problems, an embodiment of the present invention provides an impedance identification method based on a physical information neural network, comprising:

[0006] Acquire a preprocessed training data set; wherein the training data set includes operating point information and original impedance data of the modular multilevel converter;

[0007] Based on the working point information, select the corresponding control physical information to train the neural network according to the purpose of impedance identification, obtain the initial physical information neural network model, and output the training control physical information;

[0008] Training the initial physical information neural network model based on the working point information and the training control physical information, and outputting training impedance data;

[0009] Comparing the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, adjusting the initial physical information neural network model until the error is less than the set threshold, thereby obtaining an impedance identification model;

[0010] The operating point information and the selected control physical information are input into the impedance identification model to obtain the impedance output at the corresponding operating point, thereby completing the impedance identification of the modular multilevel converter.

[0011] Optionally, before obtaining the preprocessed training data set, the method further includes:

[0012] The operating point information of the modular multilevel converter and the original impedance data at different frequencies are obtained through simulation models or physical measurements to obtain a training data set.

[0013] Optionally, the preprocessing includes phase expansion and normalization of the training data set.

[0014] Further, the control physical information includes differential mode side control physical information and common mode side control physical information;

[0015] The differential mode side control physical information includes the physical information C of the differential mode control and the current dmi_p and differential mode control and voltage physical information C dmu_p ;

[0016] The common mode side control physical information includes common mode control and current physical information C cmi_p and common mode control and voltage physical information C cmu_p ;

[0017] The physical information definition of the differential mode control and current is as follows:

[0018]

[0019] The physical information definition of the differential mode control and voltage is:

[0020]

[0021] The common mode control and the physical information definition of the current are:

[0022]

[0023] The physical information definition of the common mode control and voltage is:

[0024]

[0025] In each definition, M dm_p,f＝fp and Mcm_p,f＝fp They represent the frequency f injected from the MMC AC side port respectively. p When there is a three-phase positive sequence disturbance, the disturbance obtained at the differential mode side control and the common mode side control; I p,f＝fp Indicates current disturbance.

[0026] Furthermore, the selecting corresponding control physical information according to the purpose of impedance identification includes:

[0027] If the impedance identifies the AC impedance of the modular multilevel converter, the differential mode side control physical information is selected;

[0028] If the impedance identifies the DC impedance of the modular multilevel converter, the common mode side control physical information is selected.

[0029] Furthermore, based on the working point information, the corresponding control physical information is selected to train the neural network according to the purpose of impedance identification, an initial physical information neural network model is obtained, and the training control physical information is output, including:

[0030] Calculate original control physical information according to the working point information;

[0031] According to the purpose of impedance identification, corresponding control physical information is selected to train the neural network, and the training control physical information is output;

[0032] The training control physical information is compared with the original control physical information to obtain a first training error; if the first training error is greater than or equal to a set threshold, the neural network is adjusted until the error is less than the set threshold, an initial physical information neural network model is obtained, and the final training control physical information is output.

[0033] Preferably, the impedance identification method based on physical information neural network, after completing the impedance identification of the modular multilevel converter, further comprises:

[0034] The impedance identification result is used to analyze the stability of the interaction between the modular multilevel converter and the power grid.

[0035] To solve the above technical problems, an embodiment of the present invention further provides an impedance identification system based on a physical information neural network, comprising:

[0036] A training data acquisition module, used to acquire a preprocessed training data set; wherein the training data set includes operating point information and original impedance data of the modular multi-level converter;

[0037] A first training module is used to select corresponding control physical information to train a neural network based on the working point information and the purpose of impedance identification, obtain an initial physical information neural network model, and output training control physical information;

[0038] A second training module is used to train the initial physical information neural network model based on the working point information and the training control physical information, and output training impedance data;

[0039] An impedance identification model construction module is used to compare the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, the initial physical information neural network model is adjusted until the error is less than the set threshold, thereby obtaining an impedance identification model;

[0040] The impedance identification module is used to input the operating point information and the selected control physical information into the impedance identification model, obtain the impedance output under the corresponding operating point, and complete the impedance identification of the modular multi-level converter.

[0041] An embodiment of the present invention provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, the impedance identification method based on the physical information neural network as described in any one of the above items is implemented.

[0042] An embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the impedance identification methods based on physical information neural networks as described above.

[0043] Compared with the prior art, an impedance identification method, system, device and medium based on a physical information neural network provided in an embodiment of the present invention first obtains the preprocessed working point information and original impedance data of a modular multilevel converter; then, according to the purpose of impedance identification, the corresponding control physical information is selected, and a neural network is trained based on the working point information and the selected control physical information to obtain an initial physical information neural network model, and the trained control physical information is output; then, the initial physical information neural network model is trained based on the working point information and the trained control physical information, and the trained impedance data is output; the trained impedance data is compared with the original impedance data to obtain a training error, and the initial physical information neural network model is adjusted according to the error to obtain an impedance identification model; finally, the impedance identification model is used to complete the impedance identification. The present invention can enrich the input data set of the neural network by combining the control physical information of the power electronic equipment, thereby improving the connection between the input data set and the output data volume of the neural network, and obtaining a better impedance fitting effect, and solving the problem that the impedance identification method based on the conventional neural network has obvious errors in the impedance amplitude and phase near the resonance peak when processing the power electronic equipment with multi-resonance peak characteristics; by training the physical information neural network twice, compared with the conventional neural network training method, the training accuracy of the neural network impedance modeling method is improved, so that the error of impedance identification is greatly reduced. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solution of the present invention, the drawings used in the implementation mode will be briefly introduced below. Obviously, the drawings described below are only some implementation modes of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 It is a flow chart of an impedance identification method based on physical information neural network provided by an embodiment of the present invention;

[0046] Figure 2 is another flow chart of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention;

[0047] Figure 3 A physical information neural network training process of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention;

[0048] Figure 4 It is a schematic diagram of the structure of a physical information neural network of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention;

[0049] Figure 5It is a physical information neural network training result and an error diagram of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention;

[0050] Figure 6 is a comparison diagram of the results of performing neural network impedance identification using the method provided by the embodiment of the present invention and the traditional method;

[0051] Figure 7 Error comparison diagram of the method of the embodiment of the present invention and the traditional method for neural network impedance identification;

[0052] Figure 8 A Nyquist curve and a simulated time domain waveform diagram made from the impedance identification results of an impedance identification method based on a physical information neural network provided in an embodiment of the present invention;

[0053] Fig. 9 It is a structural block diagram of an impedance identification system based on physical information neural network provided by an embodiment of the present invention;

[0054] Fig.10 It is a structural block diagram of a terminal device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. 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.

[0056] See also Figure 1 , is a flow chart of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention, the method comprising steps S1 to S5:

[0057] S1. Obtaining a preprocessed training data set; wherein the training data set includes operating point information and original impedance data of a modular multilevel converter;

[0058] Specifically, before obtaining the preprocessed training data set, the method further includes:

[0059] The operating point information of the modular multilevel converter and the original impedance data at different frequencies are obtained through simulation models or physical measurements to obtain a training data set.

[0060] Furthermore, the preprocessing includes phase expansion and normalization of the training data set.

[0061] It should be noted that the impedance and operating point information (generally including the d-axis voltage v d , q-axis voltage v q , d-axis current i d , q-axis current i q , frequency f) and other input data, so as to pre-process the data, mainly through phase expansion and normalization. Among them, phase expansion is mainly aimed at the problem that the impedance phase angle is limited to (-π,π) after using the angle function, and the impedance phase is shifted by 2π to expand the phase; normalization is to solve the dimensional difference between the input data, which can ensure that the data is within the range of (-1,1).

[0062] It is worth mentioning that the operating point information describes the performance characteristics of the converter under different operating conditions. This data can provide detailed information about the behavior and response of the converter. Once this data is obtained, you can start to build a training data set and perform phase expansion on the training data set to increase the dimension of the data; perform normalization to make the data comparable at different scales, which will help ensure that the model can learn and generalize better during the training process.

[0063] S2. Based on the working point information, select the corresponding control physical information training neural network according to the purpose of impedance identification, obtain the initial physical information neural network model, and output the training control physical information;

[0064] Specifically, the control physical information includes differential mode side control physical information and common mode side control physical information;

[0065] The differential mode side control physical information includes the physical information C of the differential mode control and the current dmi_p and differential mode control and voltage physical information C dmu_p ;

[0066] The common mode side control physical information includes common mode control and current physical information C cmi_p and common mode control and voltage physical information C cmu_p ;

[0067] The physical information definition of the differential mode control and current is as follows:

[0068]

[0069] The physical information definition of the differential mode control and voltage is:

[0070]

[0071] The common mode control and the physical information definition of the current are:

[0072]

[0073] The physical information definition of the common mode control and voltage is:

[0074]

[0075] In each definition, M dm_p,f＝fp and M cm_p,f＝fp They represent the frequency f injected from the MMC AC side port respectively. p When there is a three-phase positive sequence disturbance, the disturbance obtained at the differential mode side control and the common mode side control; I p,f＝fp Indicates current disturbance.

[0076] It should be noted that, according to the characteristics of MMC, physical information can be divided into differential mode side control physical information and common mode side control physical information:

[0077] For the differential mode side of the MMC, when a specific frequency f is injected from the AC side of the MMC to the port p When there is a three-phase positive sequence disturbance, the differential mode control can obtain the disturbance M dm_p,f＝fp , combined with the current disturbance I p,f＝fp , analogy with the definition of impedance, the relationship between differential mode control and current can be defined as C dmi_p , as shown in formula (1), characterizes the physical information of differential mode control and current, C dmi_p The same amplitude and phase angle data are included (using bold variables (such as M dm_p , C dmi_p etc.) represents the phasor, which has both amplitude and phase angle information); the relationship between differential mode control and voltage is defined as C dmu_p , as shown in formula (2), characterizes the physical information of differential mode control and voltage, C dmu_p The same includes amplitude and phase angle data.

[0078] For the common-mode side of the MMC, when a specific frequency f is injected from the MMC DC bus to the port, p When there is a single-phase disturbance, the common mode control can obtain the disturbance M cm_p,f＝fp , combined with the current disturbance I p,f＝fp , analogy with the definition of impedance, the relationship between common mode control and current can be defined as C cmi_p , as shown in formula (3), characterizing the physical information C of common mode control and current cmi_p The same includes amplitude and phase angle data; the relationship between common mode control and voltage is defined as C cmu_p , as shown in formula (4), characterizes the physical information of common mode control and voltage, C cmu_p The same includes amplitude and phase angle data.

[0079] Furthermore, the selecting corresponding control physical information according to the purpose of impedance identification includes:

[0080] If the impedance identifies the AC impedance of the modular multilevel converter, the differential mode side control physical information is selected;

[0081] If the impedance identifies the DC impedance of the modular multilevel converter, the common mode side control physical information is selected.

[0082] It should be noted that the special feature of MMC is that its AC / DC ports have two control degrees of freedom, differential mode and common mode, and the AC / DC control characteristics and AC / DC impedance characteristics of MMC need to be fully considered. Therefore, according to the equivalent AC / DC loop model of MMC, since the differential mode side controls the AC side, when choosing to identify the AC impedance of MMC, the control physical information C of the differential mode side can be selected. dmi_p and C dmu_p , while ignoring the common mode side control physical information. Similarly, since the common mode side controls the DC side, when choosing to identify the DC impedance of the MMC, the control physical information C of the common mode side can be selected. cmi_p and C cmu_p , while ignoring the physical information of differential mode control.

[0083] S3, training the initial physical information neural network model based on the working point information and the training control physical information, and outputting training impedance data;

[0084] S4, comparing the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, adjusting the initial physical information neural network model until the error is less than the set threshold, thereby obtaining an impedance identification model;

[0085] S5. Inputting the operating point information and the selected control physical information into the impedance identification model, obtaining the impedance output at the corresponding operating point, and completing the impedance identification of the modular multilevel converter.

[0086] Compared with the prior art, the embodiment of the present invention provides an impedance identification method based on a physical information neural network. By combining the control physical information of the power electronic equipment, the input data set of the neural network can be enriched, thereby improving the connection between the input data set and the output data volume of the neural network, and obtaining a better impedance fitting effect. The conventional neural network-based impedance identification method solves the problem that there are obvious errors in the impedance amplitude and phase near the resonance peak when processing power electronic equipment with multi-resonance peak characteristics; and through two physical information neural network trainings, compared with the conventional neural network training method, the training accuracy of the neural network impedance modeling method is improved, so that the error of impedance identification is greatly reduced.

[0087] In an optional embodiment, based on the working point information, the corresponding control physical information is selected to train the neural network according to the purpose of impedance identification, an initial physical information neural network model is obtained, and the training control physical information is output, including:

[0088] Calculate original control physical information according to the working point information;

[0089] According to the purpose of impedance identification, corresponding control physical information is selected to train the neural network, and the training control physical information is output;

[0090] The training control physical information is compared with the original control physical information to obtain a first training error; if the first training error is greater than or equal to a set threshold, the neural network is adjusted until the error is less than the set threshold, an initial physical information neural network model is obtained, and the final training control physical information is output.

[0091] Based on any of the above embodiments, the impedance identification method based on physical information neural network, after completing the impedance identification of the modular multilevel converter, further includes:

[0092] The impedance identification result is used to analyze the stability of the interaction between the modular multilevel converter and the power grid.

[0093] It is worth noting that, by adding the physical information of the MMC controller as the working point, the embodiment of the present invention can significantly reduce the error of MMC impedance identification compared with the conventional neural network training method, thereby providing a basis for stability analysis.

[0094] In order to more clearly describe the technical solutions in the embodiments of the present invention, see Figure 2 , is another flow chart of an impedance identification method based on a physical information neural network provided by an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0095] Step 1: Acquire data. Specifically, the sequence impedance and operating point information of the MMC at different frequencies can be acquired by frequency scanning (which can be used for subsequent calculation of the physical information of the differential mode side and common mode side control of the MMC).

[0096] Step 2: Preprocess the data (mainly phase expansion and normalization) to obtain the training data set.

[0097] Step 3: Select the control physical information. According to the characteristics of MMC, the physical information can be divided into the differential mode side control physical information and the common mode side control physical information. According to the purpose of impedance identification, if the impedance identifies the MMC AC impedance, then select the differential mode side control physical information C. dmi_p and C dmu_pIf the impedance identifies the MMC DC impedance, then select the common mode side control physical information C cmi_p and C cmu_p .

[0098] Step 4: For the first training, the working point information is used as input, and the completed control physical information is selected as output to train the neural network and obtain the control information neural network model. After the training is completed, it is necessary to determine whether the error meets the requirements (in specific implementation, it is defined as an error less than 5%). When the error meets the requirements, the required initial control physical information neural network model is obtained, otherwise the neural network is adjusted to repeat the above steps.

[0099] Step 5: The second training is based on the control physical information neural network model of step 4 to perform impedance identification on the AC side or DC side of the MMC. The training impedance data generated by the model is compared with the original impedance data to obtain the training error (in specific implementation, it is defined as an error less than 5%). When the error meets the requirements, the required neural network model (impedance identification model) is obtained, thereby completing the impedance identification. Otherwise, the neural network is adjusted to repeat the above steps.

[0100] For example, the physical information neural network training process is divided into two training processes. Figure 3 As shown. The first training neural network input is the operating point information, and the output is the amplitude and phase angle of the control physical information selected in step 3 above. Taking the neural network input as the operating point (d-axis voltage v d , q-axis voltage v q , d-axis current i d , q-axis current i q , frequency f), the output is the differential mode side control physical information C dmi_p and C dmu_p Taking the amplitude and phase angle of the first training as an example, the neural network established in the first training is a 5-input 4-output system. The second training is based on the control information generated after the first training. The neural network input is the operating point information and the control information generated in the first training. The neural network output is the amplitude and phase angle of the sequence impedance. That is, the neural network input is the operating point (d-axis voltage v d , q-axis voltage v q , d-axis current i d , q-axis current i q , frequency f), differential mode side control physical information C dmi_p and C dmu_p The amplitude and phase angle of the output is the sequence impedance Z p Taking the amplitude and phase angle as an example, the neural network established in the second training is a 9-input 2-output system.

[0101] During specific implementation, the embodiment of the present invention also verifies the impedance identification method based on the physical information neural network by building a simulation model in Matlab / Simulink. Taking the AC side of the MMC as an example, a three-phase positive sequence voltage disturbance with a frequency range of 1 to 1000 Hz and an interval of 2 Hz is injected into the AC side of the MMC. The operating point is active power P = 30MW to P = 50MW, with an interval of 2.5MW. The BP neural network is used for training, which includes an input layer, an output layer and multiple hidden layers. The layers are fully connected, and the neurons in each layer are not connected to each other. The structure is as follows Figure 4 As shown. For the physical information neural network mentioned in any of the above embodiments, when the input and output data are given, a certain neural network model can be obtained through the setting of the number of layers m, the number of layers n and the neurons. At the same time, the embodiment of the present invention also uses a traditional neural network for impedance identification for comparison. In the embodiment of the present invention, in order to ensure the consistency of the scale of the neural network, the number of hidden layers of both neural networks is 3, and the number of neurons corresponding to each layer is 10\12\15 respectively.

[0102] like Figure 5-7 The figure shows the training results and errors of the physical information neural network in the embodiment of the present invention, wherein the physical information is controlled as an intermediate bridge for training impedance, and the output after training through the first layer of the neural network has a better fitting result. Taking the selection of the AC impedance of the MMC for identification as an example, Figure 5 As shown, the control physical information C of the differential mode side of its output dmi_p The maximum amplitude error is 0.2230dB, the maximum phase angle error is 0.2719°, C dmu_p The maximum amplitude error is 0.1442dB, and the maximum phase angle error is 0.2308°. The error values ​​are small, indicating that a good fitting effect can be achieved.

[0103] Furthermore, the embodiment of the present invention also uses the fitted control physical information and working point information as the input of the second layer neural network to train the system impedance. Figure 6 , is the result of the method of the present invention provided by the embodiment of the present invention and the use of the traditional method to perform neural network impedance identification; see Figure 7 , is the error between the method provided by the embodiment of the present invention and the traditional method for neural network impedance identification. Figures 6-7As shown, at this time, the maximum error of the impedance amplitude of the method of the present invention is 0.2489dB, and the maximum error of the impedance phase angle is 0.3280°; while using the traditional neural network training method, since the impedance of the MMC is prone to mutation in the low frequency band, the actual training effect is not good. At this time, the maximum error of the impedance amplitude is 1.0794dB, and the maximum error of the impedance phase angle is 5.0632°. At this time, the maximum error of the impedance identification result based on the joint drive of control physical information and data is 0.2489dB, and the maximum error of the impedance phase angle is 0.3280°. The errors are less than 0.5%, which is 76.9% and 93.5% lower than the maximum errors of the amplitude and phase of the conventional neural network. The use of the physical information neural network training method in the embodiment of the present invention greatly reduces the impedance training error, verifying the effectiveness of the method of the present invention.

[0104] Preferably, after completing the neural network impedance identification, the impedance theory can be used to determine the system stability. That is, when the short-circuit ratio of the power grid is 1.85 and the MMC output power is rated power, the neural network impedance identification results are used to analyze the stability of the interaction between the MMC and the power grid. Figure 8 , is a Nyquist curve and a simulation time domain waveform verification of the impedance result after the impedance identification method based on physical information neural network provided by the embodiment of the present invention is used. Figure 8 As shown, when the system short-circuit ratio switches from 2.09 to 1.85 at the simulation time of 3 s, the system begins to oscillate, which also verifies the effectiveness of the impedance identification method based on physical information neural network described in the present invention.

[0105] Based on the above method embodiments, the present invention provides corresponding system embodiments.

[0106] See also Fig. 9 , is a structural block diagram of an impedance identification system based on a physical information neural network provided by an embodiment of the present invention, the system comprising:

[0107] The training data acquisition module 21 is used to acquire a preprocessed training data set; wherein the training data set includes the operating point information and the original impedance data of the modular multi-level converter;

[0108] A first training module 22 is used to select corresponding control physical information to train a neural network based on the working point information and the purpose of impedance identification, obtain an initial physical information neural network model, and output training control physical information;

[0109] A second training module 23, used for training the initial physical information neural network model based on the working point information and the training control physical information, and outputting training impedance data;

[0110] The impedance identification model building module 24 is used to compare the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, the initial physical information neural network model is adjusted until the error is less than the set threshold, thereby obtaining an impedance identification model;

[0111] The impedance identification module 25 is used to input the operating point information and the selected control physical information into the impedance identification model, obtain the impedance output under the corresponding operating point, and complete the impedance identification of the modular multi-level converter.

[0112] Preferably, the impedance identification system based on physical information neural network further includes a data acquisition module for acquiring the operating point information of the modular multi-level converter and the original impedance data at different frequencies through a simulation model or physical measurement to obtain a training data set.

[0113] Preferably, the first training module 22 comprises a control physical information selection unit, which is used to:

[0114] If the impedance identifies the AC impedance of the modular multilevel converter, the differential mode side control physical information is selected;

[0115] If the impedance identifies the DC impedance of the modular multilevel converter, the common mode side control physical information is selected.

[0116] Preferably, the first training module 22 further includes a first training unit, which is used to:

[0117] Calculate original control physical information according to the working point information;

[0118] According to the purpose of impedance identification, corresponding control physical information is selected to train the neural network, and the training control physical information is output;

[0119] The training control physical information is compared with the original control physical information to obtain a first training error; if the first training error is greater than or equal to a set threshold, the neural network is adjusted until the error is less than the set threshold, an initial physical information neural network model is obtained, and the final training control physical information is output.

[0120] Preferably, the impedance identification system based on physical information neural network further includes a stability analysis module for:

[0121] The impedance identification result is used to analyze the stability of the interaction between the modular multilevel converter and the power grid.

[0122] It should be noted that an impedance identification system based on a physical information neural network provided in an embodiment of the present invention is used to execute all process steps of an impedance identification method based on a physical information neural network in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be elaborated upon.

[0123] The embodiment of the present invention also provides a terminal device, such as Fig.10 , which is a structural block diagram of a preferred embodiment of a terminal device provided by the present invention. The terminal device includes a processor 31, a memory 32, and a computer program stored in the memory 32 and configured to be executed by the processor 31. When the processor 31 executes the computer program, the impedance identification method based on the physical information neural network as described in any of the above embodiments is implemented.

[0124] In addition, an embodiment of the present invention further provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the impedance identification method based on the physical information neural network as described in any of the above embodiments.

[0125] When the processor 31 executes the computer program, the steps in the embodiment of the impedance identification method based on physical information neural network are implemented, for example: Figure 1 Alternatively, when the processor 31 executes the computer program, the functions of each module in the embodiment of the impedance identification system based on physical information neural network are realized, for example Fig. 9 The functions of each module of the impedance identification system based on physical information neural network are shown.

[0126] Preferably, the computer program can be divided into one or more modules / units, which are stored in the memory 32 and executed by the processor 31 to implement the present invention. The one or more modules / units can be a series of computer program instruction segments that can implement specific functions, and the instruction segments are used to describe the execution process of the computer program in the terminal device.

[0127] The processor 31 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor, or the processor 31 can be any conventional processor. The processor 31 is the control center of the terminal device, and various parts of the terminal device are connected using various interfaces and lines.

[0128] The memory 32 mainly includes a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function, etc., and the data storage area can store related data, etc. In addition, the memory 32 can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash card (Flash Card), etc., or the memory 32 can also be other volatile solid-state storage devices.

[0129] It should be noted that the above terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will understand that Fig.10 The structural block diagram shown is only a structural example of the above-mentioned terminal device and does not constitute a structural limitation of the above-mentioned terminal device. The above-mentioned terminal device may include more or fewer components than shown in the figure, or combine certain components, or different components.

[0130] In summary, an impedance identification method, system, device and medium based on a physical information neural network provided by an embodiment of the present invention first obtains the preprocessed working point information and original impedance data of a modular multilevel converter; then, according to the purpose of impedance identification, the corresponding control physical information is selected, and a neural network is trained based on the working point information and the selected control physical information to obtain an initial physical information neural network model, and the trained control physical information is output; then, the initial physical information neural network model is trained based on the working point information and the trained control physical information, and the trained impedance data is output; the trained impedance data is compared with the original impedance data to obtain a training error, and the initial physical information neural network model is adjusted according to the error to obtain an impedance identification model; finally, the impedance identification model is used to complete the impedance identification. The present invention can enrich the input data set of the neural network by combining the control physical information of the power electronic equipment, thereby improving the connection between the input data set and the output data volume of the neural network, and obtaining a better impedance fitting effect, and solving the problem that the impedance identification method based on the conventional neural network has obvious errors in the impedance amplitude and phase near the resonance peak when processing the power electronic equipment with multi-resonance peak characteristics; by training the physical information neural network twice, compared with the conventional neural network training method, the training accuracy of the neural network impedance modeling method is improved, so that the error of impedance identification is greatly reduced.

[0131] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. An impedance identification method based on physical information neural network, characterized in that: include: Acquire a preprocessed training data set; wherein the training data set includes operating point information and original impedance data of the modular multilevel converter; Based on the working point information, select the corresponding control physical information to train the neural network according to the purpose of impedance identification, obtain the initial physical information neural network model, and output the training control physical information; Training the initial physical information neural network model based on the working point information and the training control physical information, and outputting training impedance data; Comparing the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, adjusting the initial physical information neural network model until the error is less than the set threshold, thereby obtaining an impedance identification model; The operating point information and the selected control physical information are input into the impedance identification model to obtain the impedance output at the corresponding operating point, thereby completing the impedance identification of the modular multilevel converter.

2. The impedance identification method based on physical information neural network according to claim 1, characterized in that: Before obtaining the preprocessed training data set, the method further includes: The operating point information of the modular multilevel converter and the original impedance data at different frequencies are obtained through simulation models or physical measurements to obtain a training data set.

3. The impedance identification method based on physical information neural network according to claim 1, characterized in that: The preprocessing includes phase expansion and normalization of the training data set.

4. The impedance identification method based on physical information neural network according to claim 1, characterized in that: The control physical information includes differential mode side control physical information and common mode side control physical information; The differential mode side control physical information includes the physical information C of the differential mode control and the current dmi_p and differential mode control and voltage physical information C dmu_p ; The common mode side control physical information includes common mode control and current physical information C cmi_p and common mode control and voltage physical information C cmu_p ; The physical information definition of the differential mode control and current is as follows: The physical information definition of the differential mode control and voltage is: The common mode control and the physical information definition of the current are: The physical information definition of the common mode control and voltage is: In each definition, M dm_p,f＝fp and M cm_p,f＝fp They represent the frequency f injected from the MMC AC side port respectively. p When there is a three-phase positive sequence disturbance, the disturbances obtained at the differential mode side control and the common mode side control; I p,f＝fp Indicates current disturbance.

5. The impedance identification method based on physical information neural network according to claim 4, characterized in that: The selecting corresponding control physical information according to the purpose of impedance identification includes: If the impedance identifies the AC impedance of the modular multilevel converter, the differential mode side control physical information is selected; If the impedance identifies the DC impedance of the modular multilevel converter, the common mode side control physical information is selected.

6. The impedance identification method based on physical information neural network according to claim 5, characterized in that: Based on the working point information, the corresponding control physical information is selected to train the neural network according to the purpose of impedance identification, an initial physical information neural network model is obtained, and the training control physical information is output, including: Calculate original control physical information according to the working point information; According to the purpose of impedance identification, corresponding control physical information is selected to train the neural network, and the training control physical information is output; The training control physical information is compared with the original control physical information to obtain a first training error; if the first training error is greater than or equal to a set threshold, the neural network is adjusted until the error is less than the set threshold, an initial physical information neural network model is obtained, and the final training control physical information is output.

7. The impedance identification method based on physical information neural network according to claim 1, characterized in that: After completing the impedance identification of the modular multilevel converter, the method further includes: The impedance identification result is used to analyze the stability of the interaction between the modular multilevel converter and the power grid.

8. An impedance identification system based on physical information neural network, characterized in that: include: A training data acquisition module, used to acquire a preprocessed training data set; wherein the training data set includes operating point information and original impedance data of the modular multi-level converter; A first training module is used to select corresponding control physical information to train a neural network based on the working point information and the purpose of impedance identification, obtain an initial physical information neural network model, and output training control physical information; A second training module is used to train the initial physical information neural network model based on the working point information and the training control physical information, and output training impedance data; An impedance identification model construction module is used to compare the training impedance data with the original impedance data to obtain a second training error; if the second training error is greater than or equal to a set threshold, the initial physical information neural network model is adjusted until the error is less than the set threshold, thereby obtaining an impedance identification model; The impedance identification module is used to input the operating point information and the selected control physical information into the impedance identification model, obtain the impedance output under the corresponding operating point, and complete the impedance identification of the modular multi-level converter.

9. A terminal device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, it implements the impedance identification method based on physical information neural network as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the impedance identification method based on physical information neural network as described in any one of claims 1 to 7.

Images