New energy unit electromechanical transient parameter identification and evaluation method fusing ANN and HCR

Abstract

The invention provides a new energy unit electromechanical transient parameter identification and evaluation method fusing ANN and HCR, and the method comprises the following steps: reading all high and low voltage ride through test data of the actual measurement or electromagnetic transient simulation of a new energy unit; basic wave positive sequence voltage, active power, reactive power, active current, reactive current and other components of all fault ride-through data are extracted based on discrete full-wave Fourier coefficients, new energy electromechanical model parameters are identified based on a neural network (ANN) and a historical case library (HCR), and new energy unit fault ride-through is simulated by PSD-BPA electromechanical transient simulation software. The method comprises the following steps: firstly, testing data and electromechanical simulation data are acquired, then fault intervals of the testing data and the electromechanical simulation data are divided based on a second-order difference adaptive algorithm, the average deviation, the average absolute deviation, the maximum deviation and the weighted average deviation of the fault intervals are calculated, finally, a two-dimensional graph and deviation calculation results are output, and the calculation results are evaluated.

Description

Technical Field

[0001] The present application relates to the technical field of new energy simulation, and in particular to a method for identifying and evaluating electromechanical transient parameters of new energy units by integrating ANN and HCR. Background Art

[0002] With the construction of new power systems and the large-scale integration of renewable energy, the proportion of renewable energy is increasing. New energy electromechanical simulation is a bridge between theoretical research and practical application. Electromechanical simulation is one of the means of grid stability analysis. The accuracy of the renewable energy model and the values ​​of the key model parameters directly affect the simulation results. Therefore, correctly identifying the fault ride-through control mode and parameters of renewable energy units is of great significance for conducting stability analysis of power grids with a high proportion of renewable energy. At present, the following problems exist in the new energy electromechanical simulation: (1) The existing recording equipment for fault ride-through test of new energy units uses its own method to extract the positive sequence component. The national standard has required the use of the full-wave Fourier method to extract the positive sequence component. The existing fault interval division only depends on the fixed time window before and after the voltage enters the fault ride-through threshold. The response characteristics of new energy units vary greatly. The fixed time window method cannot accurately identify the transient and steady-state intervals during the fault ride-through period of the new energy unit.

[0003] (2) The existing new energy electromechanical model parameter identification mainly relies on manual identification. Manual identification has the problems of relying on human subjectivity and experience, as well as low efficiency and accuracy. In addition, the on-site fault ride-through test is divided into 120 groups of test data such as different voltage drops / rises and different operating powers. The workload of manual identification is large and cannot take into account all working conditions, resulting in large deviations in some working conditions.

[0004] (3) Existing electromechanical identification software has problems in data compatibility and waveform recognition; and electromechanical transient simulation modeling is required for each new energy station added each year, and the electromechanical modeling must be completed within 6 months of grid connection, which is a huge workload. Summary of the Invention

[0005] The embodiments of the present application provide a method for identifying and evaluating electromechanical transient parameters of new energy units by integrating ANN and HCR to solve the above technical problems.

[0006] In view of this, the present application provides a method for identifying and evaluating electromechanical transient parameters of new energy units by integrating ANN and HCR, which is characterized by comprising the following steps: S1. Obtain the instantaneous value data of three-phase voltage and current of the new energy unit; S2. extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on discrete full-wave Fourier coefficients; S3. Based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, identify the high voltage and low voltage ride-through active control mode, reactive control mode and transient parameters of the new energy unit through a combination of an artificial neural network (ANN) and a historical case library (HCR); S4. Based on the transient parameters of the new energy unit, use the PSD-BPA new energy electromechanical transient simulation software to simulate the high voltage and low voltage ride-through capabilities of the new energy unit to obtain the fundamental positive sequence component data of the new energy electromechanical simulation; S5. Dividing the fault intervals of the fundamental positive sequence component data of the high voltage and low voltage ride through test of the new energy generator set and the fundamental positive sequence component data of the new energy electromechanical simulation based on a second-order difference adaptive algorithm; S6. Calculate the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fundamental positive sequence component data of the high voltage and low voltage ride-through test data of the new energy unit and the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation, obtain the calculation results, and evaluate the calculation results.

[0007] Optionally, the fundamental positive sequence component includes fundamental positive sequence voltage, active power, reactive power, active current and reactive current components.

[0008] Optionally, obtaining the three-phase voltage and current instantaneous value data of the new energy generator set in step S1 specifically includes: The high voltage and low voltage ride-through performance of new energy units are tested on-site and in the laboratory using fault ride-through detection equipment, and the instantaneous three-phase voltage and current data on the high voltage side of the new energy unit transformer or the new energy unit terminal are read.

[0009] Optionally, in step S2, the fundamental positive sequence component of the three-phase voltage and current instantaneous value data is extracted based on the discrete full-wave Fourier coefficient, specifically comprising: based on the three-phase voltage and current instantaneous value data, using the discrete full-wave Fourier coefficient method to calculate the three-phase voltage and current Fourier coefficients of the fundamental component within 20ms one by one, and obtaining the fundamental positive sequence voltage component for:

[0010] Where, is the sinusoidal component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence voltage; Fundamental positive sequence active power component for:

[0011] Where, is the cosine component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence current, is the sinusoidal component of the fundamental positive sequence voltage, is the sinusoidal component of the fundamental positive sequence current; Fundamental positive sequence reactive power component for:

[0012] Where, is the cosine component of the fundamental positive sequence voltage, is the sinusoidal component of the fundamental positive sequence current, is the sinusoidal component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence current; Fundamental positive sequence active current component for:

[0013] Where, is the fundamental positive sequence active power component, is the fundamental positive sequence voltage component; Fundamental positive sequence reactive current component for:

[0014] Where, is the fundamental positive sequence reactive power component, is the fundamental positive sequence voltage component.

[0015] Optionally, in step S3, based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, the active control mode, reactive control mode and transient parameters of the high voltage and low voltage ride-through of the new energy unit are identified by combining a neural network with a historical case library, specifically including: manually identifying the new energy electromechanical transient parameters based on the voltage drop to 0%, 20%, 35%, 50%, 75% single-phase fault, two-phase fault, three-phase fault, and the voltage rise to 115%, 120%, 130%Un two-phase fault, three-phase fault, and establishing a historical case library of electromechanical transient parameters of the new energy unit; the historical case library contains the fundamental wave before, during and after the fault Positive-sequence voltage, active power, reactive power, active current, and reactive current component information; the fundamental positive-sequence voltage, active power, reactive power, active current, and reactive current component information before, during, and after the fault in the historical case library is used as the input layer of the neural network; then, based on the fundamental positive-sequence components of the three-phase voltage and current instantaneous value data, the neural network reads the fundamental positive-sequence voltage, active power, reactive power, active current, and reactive current components of the high-voltage and low-voltage ride-through test data before, during, and after the fault information, and outputs the identified high-voltage and low-voltage ride-through active control mode, reactive control mode, and transient parameters of the new energy unit.

[0016] Optionally, the process of solving the second-order difference of the fundamental positive-sequence active current component and the fundamental positive-sequence reactive current component data of the new energy generator in step S5 specifically includes: Assume that the data set of the fundamental positive sequence active current component of the new energy generator set is for: , Where, The first, second and nth data of the fundamental positive sequence active current component of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence active current component of the new energy unit. for: ; Assume that the array of fundamental positive sequence reactive current of new energy generator is for: , Where, The first, second and nth data of the fundamental positive sequence reactive current of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence reactive current component of the new energy unit. for: .

[0017] Optionally, in step S5, the fault interval is: the wind turbine is divided into three intervals: pre-fault interval, fault interval, and post-fault interval; the photovoltaic power generation unit and the energy storage system are divided into five intervals: pre-fault interval, fault transient interval, fault steady-state interval, post-fault transient interval, and post-fault steady-state interval.

[0018] Optionally, the pre-fault interval is: an interval of 20 ms before the voltage drops below 0.9 and rises above 1.1; The fault interval is: the 20ms between the voltage dropping below 0.9 and the voltage rising above 1.1 is the starting point t B1start The first 20ms between the voltage rising above 0.9 and the voltage returning to below 1.1 is the end point t B2end interval; The fault steady-state interval is: After determining the fault interval, at t B1start With t B2end Calculate the second-order difference of the fundamental positive sequence reactive current component data within the time interval The end point of the fault transient interval is less than 0.05. B1end , then t B1end With t B2end The time interval is the fault steady-state interval; The post-fault steady-state interval is: After determining the fault interval, t B2endThe interval after the time is the post-fault interval, and the second-order difference of the fundamental positive sequence active current component is calculated in the post-fault interval The end point of the transient interval after the fault is less than 0.05. C1end ; then t C1end The time interval after the fault is the post-fault steady-state interval.

[0019] Optionally, in step S6, the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation mainly refer to the NB / T 31053-2021 "Verification Procedure for Electrical Simulation Models of Wind Turbines" standard, GB / T 32892-2016 "Photovoltaic Power Generation System Model and Parameter Testing Procedure" standard, and GB / T 44117-2024 "Electrochemical Energy Storage Power Station Model Parameter Testing Procedure" standard.

[0020] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages: The present application provides a method for identifying and evaluating electromechanical transient parameters of new energy units by integrating ANN and HCR, which includes steps S1, obtaining the three-phase voltage and current instantaneous value data of the new energy unit; S2, extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on discrete full-wave Fourier coefficients; S3, based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, identifying the high voltage and low voltage ride-through active control mode, reactive control mode and transient parameters of the new energy unit through a neural network combined with a historical case library; S4, based on the transient parameters of the new energy unit, simulating the high voltage and low voltage ride-through capability of the new energy unit using PSD-BPA new energy electromechanical transient simulation software to obtain the fundamental positive sequence component data of the new energy electromechanical simulation; S5, dividing the fundamental positive sequence component data of the high voltage and low voltage ride-through test of the new energy unit and the new energy electromechanical simulation based on a second-order difference adaptive algorithm. The fault interval of the fundamental positive sequence component data; S6, calculate the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval of the fundamental positive sequence component data of the high voltage and low voltage ride-through test data of the new energy unit and the fundamental positive sequence component data of the new energy electromechanical simulation, obtain the calculation results, and evaluate the calculation results; solve the problems of large positive sequence extraction deviation and diverse recording formats of the on-site new energy unit fault ride-through recording equipment, and use the discrete full-wave Fourier algorithm to extract the fundamental positive sequence voltage, active power, reactive power, active current, reactive current, etc. of the measured data, and the accuracy of the data fundamental positive sequence component extraction is increased to 100%; greatly improve work efficiency; by calculating the average deviation, average absolute deviation, maximum deviation and weighted average absolute deviation of the fault interval of the measured and simulated fault ride-through data, the electromechanical model parameters in each fault interval are judged to be qualified through the evaluation criteria. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] In order to more clearly express the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0022] Figure 1 This is a step diagram of the electromechanical transient parameter identification and evaluation method of a new energy generator set integrating ANN and HCR provided in an embodiment of the present application; Figure 2 Schematic diagram of fault interval division of the electromechanical transient parameter identification and evaluation method of a new energy generator set integrating ANN and HCR provided in an embodiment of the present application; Figure 3 Part of the low voltage ride-through fundamental positive sequence component diagram of the electromechanical transient parameter identification and evaluation method for a new energy generator set integrating ANN and HCR provided in an embodiment of the present application; Figure 4 A comparison chart of 20%Un low voltage ride-through measurement and electromechanical simulation of the electromechanical transient parameter identification and evaluation method for new energy units integrating ANN and HCR provided in an embodiment of the present application; Figure 5 Part of the high voltage ride-through fundamental positive sequence component diagram of the electromechanical transient parameter identification and evaluation method for a new energy generator set integrating ANN and HCR provided in an embodiment of the present application; Figure 6 This is a comparison chart between the 130%Un high voltage ride-through measurement and electromechanical simulation of the electromechanical transient parameter identification and evaluation method for new energy units integrating ANN and HCR provided in the embodiments of this application. DETAILED DESCRIPTION

[0023] The following describes the embodiments of the present invention through specific examples. 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. The details in this specification can also 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 following embodiments and features in the embodiments can be combined with each other unless they conflict.

[0024] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention. Therefore, the drawings only show the layers related to the present invention and are not drawn according to the number, shape and size ratio of the layers in actual implementation. In actual implementation, the type and number of each layer can be changed at will, and the layer layout may also be more complicated.

[0025] In the following description, numerous details are set forth to provide a more thorough explanation of the embodiments of the present invention; however, it is apparent to one skilled in the art that the embodiments of the present invention may be practiced without these specific details.

[0026] like Figures 1 to 6 FIG. 1 is a flow chart of a method for identifying and evaluating electromechanical transient parameters of a new energy generator set by integrating ANN and HCR in an embodiment of the present application. The method for identifying and evaluating electromechanical transient parameters of a new energy generator set by integrating ANN and HCR in this embodiment includes the following steps: S1. Obtain the instantaneous value data of three-phase voltage and current of the new energy unit; S2. extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on discrete full-wave Fourier coefficients; S3. Based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, identify the high voltage and low voltage ride-through active control mode, reactive control mode and transient parameters of the new energy unit through a combination of an artificial neural network (ANN) and a historical case library (HCR); S4. Based on the transient parameters of the new energy unit, use the PSD-BPA new energy electromechanical transient simulation software to simulate the high voltage and low voltage ride-through capabilities of the new energy unit to obtain the fundamental positive sequence component data of the new energy electromechanical simulation; S5. Dividing the fault intervals of the fundamental positive sequence component data of the high voltage and low voltage ride through test of the new energy generator set and the fundamental positive sequence component data of the new energy electromechanical simulation based on a second-order difference adaptive algorithm; S6. Calculate the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fundamental positive sequence component data of the new energy generator set's high voltage and low voltage ride-through test data and the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation, obtain calculation results, and evaluate the calculation results; The problem of large positive sequence extraction deviation and diverse recording formats of on-site new energy unit fault crossing recording equipment is solved. The discrete full-wave Fourier algorithm is used to extract the fundamental positive sequence voltage, active power, reactive power, active current, reactive current, etc. of the measured data. The accuracy of the fundamental positive sequence component extraction of the data is increased to 100%; the work efficiency is greatly improved; by calculating the average deviation, average absolute deviation, maximum deviation and weighted average absolute deviation of the fault interval of the measured and simulated fault crossing data, the electromechanical model parameters in each fault interval are judged to be qualified according to the evaluation criteria.

[0027] Furthermore, the fundamental positive sequence component includes fundamental positive sequence voltage, active power, reactive power, active current and reactive current components.

[0028] Furthermore, in step S1, the three-phase voltage and current instantaneous value data of the new energy generator set are obtained, specifically including: The high voltage and low voltage ride-through performance of new energy units are tested on-site and in the laboratory using fault ride-through detection equipment, and the instantaneous value data of the three-phase voltage and current at the machine end of the new energy unit are read.

[0029] Furthermore, in step S2, extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on the discrete full-wave Fourier coefficients specifically includes: Based on the three-phase voltage and current instantaneous value data, the discrete full-wave Fourier coefficient method is used to calculate the three-phase voltage and current Fourier coefficients of the fundamental component within 20ms one by one, and the fundamental positive sequence voltage component is obtained. for:

[0030] Fundamental positive sequence active power component for:

[0031] Fundamental positive sequence reactive power component for:

[0032] Fundamental positive sequence active current component for:

[0033] Fundamental positive sequence reactive current component for: .

[0034] Specifically, based on the instantaneous values ​​of the three-phase voltage and current, the discrete full-wave Fourier coefficient method is used to calculate the three-phase voltage and current Fourier coefficients of the fundamental component within 20ms. Taking the voltage and current of phase a as an example, the voltage and current Fourier coefficients at time t1 are:

[0035]

[0036]

[0037]

[0038] Where: is the cosine component of the fundamental wave of phase a voltage; is the fundamental sinusoidal component of phase a voltage; is the cosine component of the fundamental wave of phase a current; is the fundamental sinusoidal component of phase a current; is the voltage of phase a; is the a-phase current; N is the number of sampling points in the 20ms fundamental wave period; t is the time variable; t1 is the first time variable in the 20ms fundamental wave period, 、 、 The second, third, and Nth times within the 20ms fundamental wave period before time t1. That is time t1; 、 、 The voltage amplitudes of phase A corresponding to the second, third, and Nth time points within the fundamental wave period 20ms before time t1; 、 、 It is the a-phase current amplitude corresponding to the second, third, and Nth time in the fundamental wave period 20ms before time t1.

[0039] The voltage and current Fourier coefficients at time tj are:

[0040]

[0041]

[0042]

[0043] tj is the first time variable within the 20ms fundamental wave period at the time to be calculated, 、 、 are the second, third, and Nth times within the 20ms fundamental wave period before time tj, That is the time tj; 、 、 The voltage amplitude of phase a corresponding to the second, third, and Nth time in the fundamental wave period 20ms before time tj; 、 、 It is the current amplitude of phase A corresponding to the second, third and Nth time in the fundamental wave period 20ms before time tj.

[0044] The voltage calculation method for phases b and c is the same as that for phase a; the current calculation method for phases b and c is the same as that for phase a voltage; by Taking the moment as an example, the voltage and current vector components of the fundamental positive sequence component are calculated using the following formula:

[0045]

[0046]

[0047]

[0048] Where: is the cosine component of the fundamental positive sequence voltage; is the sinusoidal component of the fundamental positive sequence voltage; is the cosine component of the fundamental positive sequence current; is the sinusoidal component of the fundamental positive sequence current; 、 are the cosine components of the fundamental wave of the voltage of phase b and phase c respectively; 、 is the fundamental sinusoidal component of the voltage of phase b and phase c; 、 They are the fundamental cosine components of phase b and phase c current respectively; 、 is the fundamental sinusoidal component of the b-phase and c-phase currents; Fundamental positive sequence voltage component for:

[0049] Fundamental positive sequence active power component for:

[0050] Fundamental positive sequence reactive power component for:

[0051] Fundamental positive sequence active current component for:

[0052] Fundamental positive sequence reactive current component for:

[0053] Furthermore, in the step S3, based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, the active control mode, reactive control mode and transient parameters of the high voltage and low voltage ride-through of the new energy unit are identified by combining the neural network with the historical case library, specifically including: manually identifying the new energy electromechanical transient parameters based on the voltage drop to 0%, 20%, 35%, 50%, 75% single-phase fault, two-phase fault, three-phase fault, and the voltage rise to 115%, 120%, 130%Un two-phase fault, three-phase fault, and establishing a historical case library of electromechanical transient parameters of the new energy unit; the historical case library contains the fundamental positive sequence components before, during and after the fault The fundamental positive sequence voltage, active power, reactive power, active current and reactive current component information of the three-phase fault are used as the input layer of the neural network; based on the fundamental positive sequence components of the three-phase voltage and current instantaneous value data, the neural network reads the fundamental positive sequence voltage, active power, reactive power, active current and reactive current components of the high voltage and low voltage ride through test data before, during and after the fault, and outputs the identified high voltage and low voltage ride through active control mode, reactive control mode and transient parameters of the new energy unit.

[0054] Specifically, the neural network (ANN) algorithm mainly adopts the BP neural network algorithm, which consists of three parts: input layer, hidden layer, and output layer; Input layer: mainly extracts the fundamental positive sequence voltage, active power, reactive power, active current, reactive current component of the measured data during and after the fault; Hidden layer: Mainly divided into active control mode and reactive control mode. Active control mode mainly includes two modes: specified active power and specified active current. Reactive control range includes two modes: specified reactive power and specified reactive current. Specified active power for: , where is the active power coefficient, is the initial active power, is the active power bias coefficient; Specified active current for: , where is the voltage coefficient, is the current coefficient, is the per-unit value of the terminal voltage amplitude, is the initial active current, is the active current bias coefficient; Specified reactive power for: , where is the reactive power coefficient, is the initial reactive power, is the reactive power bias coefficient; Specified reactive current for: , where is the reactive current coefficient, is the per-unit value of the terminal voltage amplitude, is the voltage setting value, is the reactive current initial value coefficient, is the initial active current, is the reactive current bias coefficient; Output layer: mainly outputs active power control mode, reactive power control mode and their transient parameters; The historical case library is established based on a large number of manually identified key control strategies and parameters for high voltage and low voltage ride-through control of electromechanical models of new energy units; The sample attributes of the BP neural network are selected from the historical case library.

[0055] Furthermore, in step S5, the process of solving the second-order difference of the fundamental positive-sequence active current component and the fundamental positive-sequence reactive current component data of the new energy generator set specifically includes: Assume that the data set of the fundamental positive sequence active current component of the new energy generator set is for: , Where, The first, second and nth data of the fundamental positive sequence active current component of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence active current component of the new energy unit. for: ; Assume that the array of fundamental positive sequence reactive current of new energy generator is for: , Where, The first, second and nth data of the fundamental positive sequence reactive current of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence reactive current component of the new energy unit. for: .

[0056] Furthermore, in step S5, the fault interval is: the wind turbine is divided into three intervals: pre-fault interval, fault interval, and post-fault interval; the photovoltaic power generation unit and the energy storage system are divided into five intervals: pre-fault interval, fault transient interval, fault steady-state interval, post-fault transient interval, and post-fault steady-state interval.

[0057] Furthermore, the pre-fault interval is: the interval of 20 ms before the voltage drops below 0.9 and rises above 1.1; The fault interval is: the 20ms between the voltage dropping below 0.9 and the voltage rising above 1.1 is the starting point t B1start The first 20ms between the voltage rising above 0.9 and the voltage returning to below 1.1 is the end point t B2end interval; The fault steady-state interval is: After determining the fault interval, at t B1start With t B2end Calculate the second-order difference of the fundamental positive sequence reactive current component data within the time interval The end point of the fault transient interval is less than 0.05. B1end , then t B1end With t B2end The time interval is the fault steady-state interval; The post-fault steady-state interval is: After determining the fault interval, t B2end The interval after the time is the post-fault interval, and the second-order difference of the fundamental positive sequence active current component is calculated in the post-fault interval The end point of the transient interval after the fault is less than 0.05. C1end ; then t C1end The time interval after the fault is the post-fault steady-state interval.

[0058] Furthermore, in step S6, the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation mainly refer to the NB / T 31053-2021 "Verification Procedure for Electrical Simulation Models of Wind Turbines" standard, GB / T 32892-2016 "Photovoltaic Power Generation System Model and Parameter Testing Procedure" standard, and GB / T 44117-2024 "Electrochemical Energy Storage Power Station Model Parameter Testing Procedure" standard.

[0059] Specifically, the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval between the measured data and the simulated data of the wind turbine mainly refer to the NB / T 31053-2021 "Verification Procedure for Electrical Simulation Models of Wind Turbines" standard; the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval between the measured data and the simulated data of the photovoltaic power generation unit mainly refer to the GB / T 32892-2016 "Photovoltaic Power Generation System Model and Parameter Testing Procedure" standard; the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval between the measured data and the simulated data of the energy storage system mainly refer to the GB / T 44117-2024 "Electrochemical Energy Storage Power Station Model Parameter Testing Procedure" standard.

[0060] For example, the discrete full-wave Fourier coefficients are used to extract the fundamental positive sequence voltage, active power, reactive power, active current, and reactive current components of the low voltage ride-through field measured data of a 5MW wind turbine. The waveforms of 20%Un, 35%Un, and 50%Un are as follows: Figure 3 As shown in the figure, the key control strategies and parameters of low voltage ride-through control of new energy electromechanical model are identified by combining artificial neural network (ANN) with historical case library (HCR): the active power control mode during low voltage ride-through is to specify the active power current, and its voltage coefficient K V is 0, the current coefficient K I is 1.1, the active current bias coefficient IP SET is 0.22; during low voltage ride-through, the reactive power control mode is specified as reactive current, and the reactive current coefficient K Vq is 2.36, the voltage setting value is V SET is 0.9, the initial value coefficient of reactive current K I =1, reactive current bias coefficient IQ SET =0; the low voltage ride-through of the new energy unit is simulated by using the new energy electromechanical transient simulation software to obtain the fundamental positive sequence voltage, active power, reactive power, active current, and reactive current component data. Based on the second-order difference adaptive algorithm, the wind turbine unit is divided into three fault intervals of the fundamental positive sequence voltage, active current, and reactive current data of the low voltage ride-through test and the electromechanical simulation data. The comparison waveform of the 20%Un low voltage ride-through is as follows Figure 4 The average deviation, average absolute deviation, maximum deviation, and weighted average deviation are shown in Table 1.

[0061]

[0062] Table 1 The full-wave Fourier coefficients of the high voltage ride-through field measured data of a 4.5MW photovoltaic power generation unit are discretely extracted to obtain the fundamental positive sequence voltage, active power, reactive power, active current, and reactive current components. Some 115%Un, 120%Un, 125%Un, and 130%Un waveforms are as follows: Figure 5 As shown in the figure, the key control strategies and parameters of the high voltage ride-through control of the electromechanical model of the photovoltaic power generation unit are identified by combining the neural network with the historical case library: the active power control mode during the high voltage ride-through is the specified active power, and its active power coefficient K Q is 1, the active power bias coefficient P SET 0; the reactive power control mode during high voltage ride-through is to specify a fixed reactive current, which is the reactive current coefficient K Vq is 2.59, the voltage setting value VSET is 1.1, and the reactive current initial value coefficient K I =1, reactive current bias coefficient IQ SET is 0; the high voltage ride-through of the photovoltaic power generation unit is simulated using new energy electromechanical transient simulation software to obtain the fundamental positive sequence voltage, active power, reactive power, active current, and reactive current component data. Based on the second-order difference adaptive algorithm, the photovoltaic power generation unit high voltage ride-through test fundamental positive sequence voltage, active current, reactive current data and electromechanical simulation data are divided into 5 fault intervals. The comparison waveform of the 130%Un high voltage ride-through is as follows Figure 6 The average deviation, average absolute deviation, maximum deviation, and weighted average deviation are shown in Table 2.

[0063]

[0064] Table 2 The terms "first," "second," "third," "fourth," and the like (if any) in the specification of the present application and the accompanying drawings are used to distinguish similar objects and are not necessarily used to describe a particular order or precedence. It should be understood that the terms used in this manner are interchangeable where appropriate, so that the embodiments of the present application described herein can, for example, be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having," and any variations thereof, are intended to cover non-exclusive inclusions. For example, a process, method, system, product, or apparatus comprising a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such process, method, product, or apparatus.

[0065] As described above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. The electromechanical transient parameter identification and evaluation method of new energy units integrating ANN and HCR is characterized by: The following steps are involved: S1. Obtain the instantaneous value data of three-phase voltage and current of the new energy unit; S2. extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on discrete full-wave Fourier coefficients; S3. Based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, identify the high voltage and low voltage ride-through active control mode, reactive control mode and transient parameters of the new energy unit through a neural network combined with a historical case library; S4. Based on the transient parameters of the new energy unit, use the PSD-BPA new energy electromechanical transient simulation software to simulate the high voltage and low voltage ride-through capabilities of the new energy unit to obtain the fundamental positive sequence component data of the new energy electromechanical simulation; S5. Dividing the fault intervals of the fundamental positive sequence component data of the high voltage and low voltage ride through test of the new energy generator set and the fundamental positive sequence component data of the new energy electromechanical simulation based on a second-order difference adaptive algorithm; S6. Calculate the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fundamental positive sequence component data of the high voltage and low voltage ride-through test data of the new energy unit and the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation, obtain the calculation results, and evaluate the calculation results.

2. The electromechanical transient parameter identification and evaluation method for new energy units integrating ANN and HCR according to claim 1 is characterized in that: The fundamental positive sequence components include fundamental positive sequence voltage, active power, reactive power, active current and reactive current components.

3. The electromechanical transient parameter identification and evaluation method of a new energy generator set integrating ANN and HCR according to claim 2 is characterized in that: In step S1, the three-phase voltage and current instantaneous value data of the new energy generator set are obtained, which specifically includes: The high voltage and low voltage ride-through performance of new energy units are tested on-site and in the laboratory using fault ride-through detection equipment, and the instantaneous three-phase voltage and current data on the high voltage side of the new energy unit transformer or the new energy unit terminal are read.

4. The electromechanical transient parameter identification and evaluation method for new energy generator sets integrating ANN and HCR according to claim 2 is characterized in that: Extracting the fundamental positive sequence component of the three-phase voltage and current instantaneous value data based on the discrete full-wave Fourier coefficients in step S2 specifically includes: Based on the three-phase voltage and current instantaneous value data, the discrete full-wave Fourier coefficient method is used to calculate the three-phase voltage and current Fourier coefficients of the fundamental component within 20ms one by one, and the fundamental positive sequence voltage component is obtained. for: Where, is the sinusoidal component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence voltage; Fundamental positive sequence active power component for: Where, is the cosine component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence current, is the sinusoidal component of the fundamental positive sequence voltage, is the sinusoidal component of the fundamental positive sequence current; Fundamental positive sequence reactive power component for: Where, is the cosine component of the fundamental positive sequence voltage, is the sinusoidal component of the fundamental positive sequence current, is the sinusoidal component of the fundamental positive sequence voltage, is the cosine component of the fundamental positive sequence current; Fundamental positive sequence active current component for: Where, is the fundamental positive sequence active power component, is the fundamental positive sequence voltage component; Fundamental positive sequence reactive current component for: Where, is the fundamental positive sequence reactive power component, is the fundamental positive sequence voltage component.

5. The electromechanical transient parameter identification and evaluation method for new energy units integrating ANN and HCR according to claim 2 is characterized in that: In step S3, based on the fundamental positive sequence component of the three-phase voltage and current instantaneous value data, the active control mode and reactive control mode of the high voltage and low voltage ride-through of the new energy generator set and their transient parameters are identified by combining a neural network with a historical case library, specifically including: Based on the manual identification of new energy electromechanical transient parameters when the voltage drops to 0%, 20%, 35%, 50%, and 75% single-phase faults, two-phase faults, and three-phase faults, and when the voltage rises to 115%, 120%, and 130%Un two-phase faults and three-phase faults, a historical case library of electromechanical transient parameters of new energy units was established; the historical case library contains information on the fundamental positive-sequence voltage, active power, reactive power, active current, and reactive current components before, during, and after the fault; the fundamental positive-sequence voltage, active power, reactive power, active current, and reactive current component information before, during, and after the fault in the historical case library is used as the input layer of the neural network; then, based on the fundamental positive-sequence components of the three-phase voltage and current instantaneous value data, the neural network reads the information on the fundamental positive-sequence voltage, active power, reactive power, active current, and reactive current components of the high voltage and low voltage ride-through test data before, during, and after the fault, and outputs the identified active control mode, reactive control mode, and transient parameters of the high voltage and low voltage ride-through of the new energy unit.

6. The electromechanical transient parameter identification and evaluation method for new energy units integrating ANN and HCR according to claim 2 is characterized in that: The process of solving the second-order difference of the fundamental positive-sequence active current component and the fundamental positive-sequence reactive current component data of the new energy generator in step S5 specifically includes: Assume that the data set of the fundamental positive sequence active current component of the new energy generator set is for: , Where, The first, second and nth data of the fundamental positive sequence active current component of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence active current component of the new energy unit. for: ； Assume that the array of fundamental positive sequence reactive current of new energy generator is for: , Where, The first, second and nth data of the fundamental positive sequence reactive current of the new energy unit are obtained by performing the second-order difference of the fundamental positive sequence reactive current component of the new energy unit. for: 。 7. The electromechanical transient parameter identification and evaluation method for new energy generator sets integrating ANN and HCR according to claim 1 is characterized in that: In step S5, the fault intervals are: the wind turbine is divided into three intervals: pre-fault interval, fault interval, and post-fault interval; the photovoltaic power generation unit and the energy storage system are divided into five intervals: pre-fault interval, fault transient interval, fault steady-state interval, post-fault transient interval, and post-fault steady-state interval.

8. The electromechanical transient parameter identification and evaluation method for new energy generator sets integrating ANN and HCR according to claim 7 is characterized in that: The pre-fault interval is the 20ms interval between the voltage dropping below 0.9 and the voltage rising above 1.

1. The fault interval is: the 20ms between the voltage dropping below 0.9 and the voltage rising above 1.1 is the starting point t B1start The first 20ms between the voltage rising above 0.9 and the voltage returning to below 1.1 is the end point t B2end interval; The fault steady-state interval is: After determining the fault interval, at t B1start With t B2end Calculate the second-order difference of the fundamental positive sequence reactive current component data within the time interval The end point of the fault transient interval is less than 0.

05. B1end , then t B1end With t B2end The time interval is the fault steady-state interval; The post-fault steady-state interval is: After determining the fault interval, t B2end The interval after the time is the post-fault interval, and the second-order difference of the fundamental positive sequence active current component is calculated in the post-fault interval The end point of the transient interval after the fault is less than 0.

05. C1end ; then t C1end The time interval after the fault is the post-fault steady-state interval.

9. The electromechanical transient parameter identification and evaluation method for new energy generator sets integrating ANN and HCR according to claim 1 is characterized in that: In step S6, the average deviation, average absolute deviation, maximum deviation, and weighted average deviation of the fault interval of the fundamental positive sequence component data of the new energy electromechanical simulation mainly refer to the NB / T 31053-2021 "Verification Procedure for Electrical Simulation Models of Wind Turbines" standard, GB / T 32892-2016 "Photovoltaic Power Generation System Model and Parameter Test Procedure" standard, and GB / T 44117-2024 "Electrochemical Energy Storage Power Station Model Parameter Test Procedure" standard.

Images