Digital-analog hybrid simulation electromechanical transient modeling method and system of energy storage system

Abstract

The invention relates to the technical field of energy storage, and particularly discloses a digital-analog hybrid simulation electromechanical transient modeling method and system for an energy storage system, and the method comprises the steps: firstly, building a digital-analog hybrid simulation test model of the energy storage system, enabling the model circuit parameters to be consistent with the same-model energy storage system of an actual energy storage power station, and enabling the model circuit parameters to be consistent with the model circuit parameters of the actual energy storage power station; checking the correctness of the signal transmission system of the digital-analog hybrid simulation system; then, carrying out an electromechanical transient modeling working condition test of the energy storage system, recording a test result, applying regression analysis to obtain electromechanical transient model parameters of the energy storage system, and establishing an electromechanical transient simulation model of the energy storage system; and finally, comparing and analyzing test results of the electromechanical transient simulation model and the test model under typical operation conditions, and calculating errors. Based on a digital-analog hybrid simulation system, an electromechanical transient simulation model consistent with an actual energy storage system of the same model in performance can be established, and an accurate energy storage model is provided for simulation of a modern power system.

Description

Technical Field

[0001] The present invention belongs to the technical field of energy storage, and in particular relates to a digital-analog hybrid simulation electromechanical transient modeling method and system for an energy storage system. Background Art

[0002] The proportion of installed capacity and power generation of renewable energy in the power grid will become higher and higher, and the adverse impact of renewable energy on the power grid will become increasingly prominent, which will seriously endanger the safe and stable operation of the power grid with a high proportion of renewable energy. Energy storage is an effective way to solve the above problems. To understand the impact of energy storage on the transient characteristics of a power grid with a high proportion of renewable energy, it is necessary to establish a simulation model of the energy storage system that can reflect the actual operating characteristics.

[0003] The core issue of energy storage system simulation modeling is to identify the parameters of the electromechanical simulation model of the energy storage system based on the test data of the energy storage system. The electromechanical transient simulation modeling of the energy storage system includes three steps: the first step is to obtain a sufficient number of test data that is consistent with the actual situation; the second step is to identify the parameters of the expressions of active current and reactive current in the stable period of the electromechanical transient model of the energy storage system during the transient period based on the test data; the third step is to calculate the error between the simulation results of the electromechanical transient simulation model and the measured results. There are two main sources of test data. The first is field testing, and the second is digital-analog hybrid simulation semi-physical testing. The first method is labor-intensive, time-consuming, and difficult to test. The objective conditions limit it and make it unsuitable for large-scale energy storage system modeling.

[0004] No relevant technical documents on electromechanical transient modeling of energy storage systems based on hybrid digital-analog simulation were found through searching. The relevant technical document information is as follows.

[0005] Prior art document 1 (CN202310173175.1, authorization announcement date 2023-05-05) "A modeling method for an energy storage battery" provides a modeling method for an energy storage battery, including: obtaining the electrochemical impedance spectrum of the energy storage battery at different internal temperatures; establishing a first energy storage battery model and a second energy storage battery model according to the electrochemical impedance spectrum of the energy storage battery at different internal temperatures; the first energy storage battery model is an energy storage battery model in which the internal temperature of the energy storage battery is within a first temperature range; the second energy storage battery model is an energy storage battery model in which the internal temperature of the energy storage battery is within a second temperature range, and the first temperature range and the second temperature range are different.

[0006] Prior art document 2 (CN202211200159.9, authorization announcement date 2023-04-14) "Dynamic behavior modeling method of power system adapting to variable operating conditions of energy storage power stations" provides a dynamic behavior modeling method of power system adapting to variable operating conditions of energy storage power stations. First, the differential algebraic equation of the energy storage power system is constructed, and then the node injection power of the energy storage power station based on the time-varying parameters is determined according to the variable operating conditions of the energy storage power station, and then the operating point set of the energy storage power system is determined. The differential algebraic equation of the energy storage power system is linearized at different operating points to obtain a linearized model of the power system based on different operating points, and then a power system state matrix based on different operating points is constructed. Finally, the power system state matrices at different operating points are combined into a power system state matrix set to construct a dynamic model of the energy storage power system based on the power system state matrix set.

[0007] Prior art document 3 (CN202310462579.2, authorization announcement date 2023-08-11) "A large-scale energy storage charging and discharging method and model based on reinforcement learning algorithm" provides a large-scale energy storage prediction model and charging and discharging strategy based on reinforcement learning algorithm, including the following steps: S1: Establish a large-scale energy storage cluster IOT Internet of Things center management platform in the cloud; S2: Construct a simulation model M 1,n ; S3: train A3C to obtain the strategy model M2 trained by A3C; S4: obtain simulation data S* through the simulation environment database of the energy storage power station or system; S5: execute the charging and discharging strategy.

[0008] Prior art document 4 (CN202110598578.1, authorization announcement date 2022-12-23) "New energy power system hardware-in-the-loop test system and its performance verification method" discloses a new energy power system hardware-in-the-loop test system, including a battery energy storage verification system, the battery energy storage verification system execution layer is connected to the motor controller verification system and the charging and discharging verification system, the motor controller verification system is connected to the motor actuator verification system, the above systems are all connected to the vehicle control verification system electrical signal, the battery energy storage verification system, the motor controller verification system and the vehicle control verification system are all equipped with HIL, and each system is equipped with a high-voltage connector for switching power on and off and a low-voltage signal connector for switching signal on and off. A performance verification method for a new energy power system hardware-in-the-loop test system is also disclosed.

[0009] The research focus of the existing technical document 1 is on the modeling of the energy storage battery itself, and the research object is the internal characteristics of the battery; the existing technical documents 2-3 are aimed at the modeling of the energy storage system, the technical document 2 focuses on the mathematical model of the energy storage system, and the technical document 3 focuses on the research of the energy storage charging and discharging method based on big data; the technical document 4 is a study on the performance test of the energy storage system based on hardware in the loop. The above research did not conduct research on the electromechanical transient modeling of the energy storage system based on digital-analog hybrid simulation, and there are two main deficiencies: first, there is no research on the combination of the advantages of digital-analog hybrid simulation in energy storage simulation modeling and energy storage modeling; second, there is no research on the electromechanical transient modeling of the energy storage system.

[0010] In summary, based on the digital-analog hybrid simulation test, this document carries out electromechanical transient modeling of energy storage systems based on digital-analog hybrid simulation. Summary of the invention

[0011] In order to address the deficiencies in the prior art, the present invention provides a method and system for modeling electromechanical transients using a hybrid digital-analog simulation of an energy storage system. Taking into account the operating characteristics of the energy storage system and combining the advantages of hybrid digital-analog simulation in the testing field, a method for establishing an electromechanical transient model of an energy storage system using hybrid digital-analog simulation is constructed, which can establish an electromechanical transient simulation model of the energy storage system that is consistent with the actual operating conditions.

[0012] The present invention adopts the following technical solution.

[0013] The present invention provides a digital-analog hybrid simulation electromechanical transient modeling method for an energy storage system, comprising:

[0014] Step A1, collect data on the actual energy storage power station, including the energy storage system controller model and software version number, primary circuit topology and component parameters, establish a hybrid digital simulation test model and a physical model of the energy storage system, and check the signal channel between the digital simulation test model and the physical model;

[0015] Step A2, setting a fault test condition, performing a digital-analog hybrid simulation fault ride-through test on the energy storage system, recording the test results, and obtaining the electromechanical transient simulation model parameters of the energy storage system based on regression analysis;

[0016] Step A3: construct an electromechanical transient simulation model of the energy storage system, compare the operation results of the electromechanical transient simulation model of the energy storage system and the digital-analog hybrid model under different working conditions, and calculate the error between the two.

[0017] Preferably, the step A1 comprises:

[0018] Step A1.1, based on the energy storage power station information, establish a digital-analog hybrid simulation test model of the same type of energy storage system, and check the signal channels between the digital simulation model and the physical device one by one;

[0019] Step A1.2, check the signal channel of the digital-analog hybrid simulation test model. When the given value of the digital model and the measured value of the energy storage controller are inconsistent, modify the transformation ratio of the digital channel of the digital model.

[0020] Preferably, in step A1.1, the energy storage system digital-analog hybrid simulation test model includes a primary circuit model module and an energy storage system control model module.

[0021] The primary circuit model module includes a circuit model unit and an energy storage system body model unit;

[0022] The energy storage system control model module includes controller hardware consistent with the field controller model and a control program version consistent with the field controller control program.

[0023] Preferably, the circuit model unit includes a power grid model, a transformer model and output end inductor and capacitor elements, and the energy storage system body model unit includes a battery model, a converter model and a DC side model.

[0024] Preferably, the step A2 comprises:

[0025] Step A2.1, setting the electromechanical transient modeling test conditions of the energy storage system, including the grid strength, the output power level of the energy storage system, the output voltage change amplitude and duration, and completing the digital-analog hybrid simulation fault ride-through test of the energy storage system under different working conditions;

[0026] Step A2.2, according to the hybrid digital-analog simulation fault ride-through test results of the energy storage system, a regression analysis method based on gradient descent is used to obtain the electromechanical transient simulation model parameters of the energy storage system;

[0027] The gradient descent method is used to realize the multivariate regression of the parameters of the electromechanical transient model of the energy storage system, and multiple influencing factors X = [x1, x2, ..., x n ] and the dependent variable Y=[y1,y2,...,y m ], the weight relationship between them is:

[0028] y k =ω0+ω1x1+ω2x2+...+ω n x n

[0029] In the formula, ω i represents the weight of the i-th influencing factor;

[0030] X=[x1,x2,...,x n ] represents the test results of the energy storage system digital-analog hybrid simulation, including the voltage in the steady-state period during the fault and the reactive power before the fault. Y = [y1, y2, ..., ym ] represents the active current amplitude and reactive current amplitude in the steady-state period during the fault period;

[0031] The vector calculation expression of the multivariate linear regression model is:

[0032] y k =ω T X

[0033] In the formula, ω T represents the weight factor, ω=[ω0,ω1,ω2,...,ω n ];

[0034] The loss function of multivariate linear regression uses mean square error, which is expressed as:

[0035]

[0036] The condition for the loss function to be minimum is that the derivative of the loss function with respect to the weight factor is 0, and the calculation expression is:

[0037]

[0038] The normal equation of the multiple regression equation is obtained as follows:

[0039] ω=(X T X) -1 X T Y

[0040] The gradient descent method is used to solve the problem, and the loss function is used to derive the weight coefficient.

[0041]

[0042] The iterative calculation expression of the weight factor based on the gradient descent algorithm is:

[0043]

[0044] In the formula, η represents the learning rate;

[0045] By iteratively updating the gradient and weight factor, the weight factor optimization result is finally obtained. The weight factor optimization result is the coefficient of the function expression of the active current and reactive current with the pre-fault current and the voltage amplitude during the fault period during the steady-state period of voltage change; by controlling the voltage drop amplitude step size, m <n。

[0046] Preferably, the step A2.1 includes the following requirements for the output power and voltage change of the energy storage system:

[0047] (1) Active power: The operating state includes charging state and discharging state, and the output power range includes 0.1P N ≤P≤0.3PN , 0.4P N ≤P≤0.7P N 、P>0.8P N Three types;

[0048] (2) Voltage drop / increase setting: The voltage drop amplitude is in steps of 0.1 pu, and the voltage drop range is 0.1 pu-0.9 pu; the voltage increase amplitude is in steps of 0.05 pu, and the voltage increase range is 1.1 pu-1.3 pu;

[0049] (3) Entering / Exiting Fault Voltage Ride-Threshold: The energy storage system enters / exits the low voltage ride-through voltage threshold test, with the voltage drop in steps of 0.01 pu and a voltage drop range of 0.8 pu-0.9 pu. The energy storage system enters / exits the high voltage ride-through voltage threshold test, with the voltage increase in steps of 0.01 pu and a voltage drop range of 1.1 pu-1.2 pu.

[0050] Preferably, step A3 comprises:

[0051] Step A3.1, calculating the electromechanical transient model error of the energy storage system, including the voltage, active power, reactive power, and reactive current of the output port of the energy storage system;

[0052] Step A3.2, the error types include the following 4 types:

[0053] (1) Average deviation

[0054]

[0055] In the formula, σ me represents the average deviation, the subscript sim represents the model simulation result, the subscript mea represents the digital-analog hybrid simulation test result, and N represents the number of test data;

[0056] (2) Maximum deviation

[0057] σ mxe =max{|x sim (i)-x mea (i)|,i=1,2,...,N}

[0058] In the formula, σ mxe represents the maximum deviation;

[0059] (3) Weighted average deviation

[0060] σ G_me =0.1σ me_pre +0.6σ me_fault +0.1σ me_post

[0061] In the formula, σ G_me represents the weighted mean deviation, σ me_pre , σ me_fault , σ me_post They represent the average deviation before, during, and after the fault, respectively;

[0062] (4) Weighted maximum deviation

[0063] σ G_mxe =0.1σ mxe_pre +0.6σ mxe_fault +0.1σ mxe_post

[0064] In the formula, σ G_mxe represents the weighted mean deviation, σ mxe_pre , σ mxe_fault , σ mxe_post They represent the average deviation before, during, and after a fault, respectively.

[0065] The present invention also provides a digital-analog hybrid simulation electromechanical transient modeling system for an energy storage system, the system being a system used by the digital-analog hybrid simulation electromechanical transient modeling method for the aforementioned energy storage system, comprising: an energy storage system performance test module, an energy storage system fault ride-through parameter identification module, and an energy storage system fault ride-through model error calculation module;

[0066] The energy storage system performance test module is used to test the energy storage system response characteristics under different grid strengths, different energy storage operating states, and different voltage drop / rise types and depths, including active power, reactive power, and reactive current;

[0067] The energy storage system fault ride-through parameter identification module is used to identify the function expression parameters between the active current, reactive current and voltage, reactive power of the energy storage system during the fault voltage ride-through period;

[0068] The energy storage system fault ride-through model error calculation module is used to calculate the error between the energy storage system electromechanical transient model simulation result and the digital-analog hybrid simulation test result, and the comparison indicators include voltage and power.

[0069] The present invention also provides a terminal, including a processor and a storage medium; the characteristics are:

[0070] The storage medium is used to store instructions;

[0071] The processor is configured to operate according to the instructions to execute the steps according to the aforementioned method.

[0072] The present invention also provides a computer-readable storage medium on which a computer program is stored, characterized in that the program implements the steps of the above method when executed by a processor.

[0073] The beneficial effect of the present invention is that, compared with the prior art,

[0074] 1. According to different grid strengths and energy storage operating conditions, a set of test conditions that meet the modeling parameter identification requirements is constructed to provide sufficient and reliable test data for energy storage system modeling;

[0075] 2. To address the black-box problem of energy storage control strategies, the voltage thresholds for energy storage systems to enter / exit low voltage ride-through and high voltage ride-through are identified by reducing the step size of the voltage drop amplitude;

[0076] 3. The topology structure, component parameters, controller type, and control program software version information of the digital and analog test system of the energy storage system are consistent with those on site, and the test results are more in line with the actual operating characteristics;

[0077] 4. A parameter identification method for the electromechanical transient simulation model of energy storage system is proposed based on multivariate regression to improve the accuracy of the electromechanical transient model of energy storage system. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1 It is a flowchart of the steps of the electromechanical transient modeling method of the energy storage system based on digital-analog hybrid simulation of the present invention;

[0079] Figure 2 It is a schematic diagram of the comparison results of the high power, discharge, and two-phase drop to 25%Un curves of the present invention;

[0080] Figure 3 It is a schematic diagram of comparison results of high power, charging, and three-phase drop to 40% Un curves of the present invention;

[0081] Figure 4 It is a schematic diagram of the low voltage ride through requirement curve of the energy storage system of the present invention. DETAILED DESCRIPTION

[0082] In order to make the purpose, technical scheme and advantages of the present invention clearer, the technical scheme of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, other embodiments obtained by ordinary technicians in this field without creative work are all within the scope of protection of the present invention.

[0083] like Figure 1 As shown, embodiment 1 of the present invention provides a digital-analog hybrid simulation electromechanical transient modeling method and system for an energy storage system, and the specific steps are as follows:

[0084] The step A1 comprises:

[0085] Based on the information of the energy storage power station, a digital-analog hybrid simulation test model of the same type of energy storage system is established, and the signal channels between the digital simulation model and the physical device are checked one by one.

[0086] In step A1.1, the energy storage system digital-analog hybrid simulation test model mainly includes two parts.

[0087] The first part is the primary circuit model, which is divided into two aspects. The first aspect is the circuit model, including the power grid model, transformer model, and output inductor and capacitor components. The second aspect is the energy storage system model, including the battery model, converter model, and DC side model.

[0088] The second part is the energy storage system control model, which is divided into two aspects. The first aspect is the controller hardware that is consistent with the on-site controller model, and the other aspect is the control program version that is consistent with the on-site controller of the same model. The control program is injected into the controller hardware during modeling.

[0089] In step A1.2, the signal channel of the digital-analog hybrid simulation system is checked to ensure the consistency of the given value of the digital model and the measured value of the energy storage controller for the same signal.

[0090] The step A2 comprises:

[0091] like Figure 4 As shown, step A2.1 sets the electromechanical transient modeling test conditions of the energy storage system, including the grid strength, the output power level of the energy storage system, the output voltage change amplitude and duration, and completes the energy storage system performance test under different conditions. The test conditions need to meet three conditions. First, they need to be consistent with the requirements of the national standards related to the grid-related performance testing of energy storage systems; second, they need to increase the number of test conditions according to the model parameter regression analysis requirements, including active and reactive output values ​​and voltage drop / increase amplitudes; third, they need to be able to identify the voltage thresholds for the energy storage system to enter / exit low voltage ride-through and high voltage ride-through.

[0092] The output power and voltage change requirements of the energy storage system are as follows:

[0093] (1) Active power: The operating state should include the charging state and the discharging state, and the output power range should include 0.1P N ≤P≤0.3P N , 0.4P N ≤P≤0.7P N 、P>0.8P N Three types;

[0094] (2) Voltage drop / rise setting: The voltage drop amplitude is in steps of 0.1pu, and the voltage drop range is 0.1pu-0.9pu; the voltage rise is in steps of 0.05pu, and the voltage rise range is 1.1pu-1.3pu. The voltage drop duration is calculated based on the low voltage ride-through requirement curve of the energy storage system.

[0095] (3) Entering / Exiting Fault Voltage Ride-Threshold: The energy storage system enters / exits the low voltage ride-through voltage threshold test, with the voltage drop in steps of 0.01 pu and a voltage drop range of 0.8 pu-0.9 pu. The energy storage system enters / exits the high voltage ride-through voltage threshold test, with the voltage increase in steps of 0.01 pu and a voltage drop range of 1.1 pu-1.2 pu.

[0096] Step A2.2, according to the mixed simulation test results of the energy storage system, a regression analysis method based on gradient descent is used to obtain the electromechanical transient simulation model parameters of the energy storage system.

[0097] The gradient descent method is used to realize the multivariate regression of the parameters of the electromechanical transient model of the energy storage system, and multiple influencing factors X = [x1, x2, ..., x n ] and the dependent variable Y=[y1,y2,...,y m ], the model is as shown in formula (1).

[0098] y k =ω0+ω1x1+ω2x2+...+ω n x n (1)

[0099] In the formula, ω i Represents the weight of the i-th influencing factor.

[0100] In the present invention, X=[x1,x2,...,x n ] represents the test results of the energy storage system digital-analog hybrid simulation, including the voltage in the steady-state period during the fault and the reactive power before the fault. Y = [y1, y2, ..., y m ] represents the active current amplitude and reactive current amplitude in the steady-state period during the fault.

[0101] The vector calculation expression of the multivariate linear regression model is:

[0102] y k =ω T X (2)

[0103] In the formula, ω T represents the weight factor, ω=[ω0,ω1,ω2,...,ω n ].

[0104] The loss function of multivariate linear regression uses mean square error, which is expressed as:

[0105]

[0106] The condition for the loss function to be minimum is that the derivative of the loss function with respect to the weight factor is 0, and the calculation expression is:

[0107]

[0108] The normal equation of the multiple regression equation is obtained as follows:

[0109] ω=(X T X) -1 X T Y (5)

[0110] In order to obtain the optimal solution, this document uses the gradient descent method. First, the loss function is derived with respect to the weight coefficient, and the result is:

[0111]

[0112] The iterative calculation expression of the weight factor based on the gradient descent algorithm is:

[0113]

[0114] In the formula, η represents the learning rate.

[0115] According to formula (7), the gradient and weight factor are continuously updated through iteration, and finally the weight factor optimization result is obtained.

[0116] In the present invention, the optimization result is the coefficients of the function expression of the active current and reactive current, the current before the fault, and the voltage amplitude during the fault period during the steady-state period of voltage change.

[0117] Note that in the present invention, by controlling the voltage drop amplitude step, m <n。

[0118] The step A3 comprises:

[0119] Step A3.1, based on the model simulation results and test results, calculate the electromechanical transient model error of the energy storage system, including voltage, active power, reactive power, reactive current, etc.;

[0120] Step A4.2, the error types include the following 4 types:

[0121] (1) Average deviation

[0122]

[0123] In the formula, σ merepresents the average deviation, the subscript sim represents the model simulation result, the subscript mea represents the digital-analog mixed simulation test result, and N represents the number of test data.

[0124] (2) Maximum deviation

[0125] σ mxe =max{|x sim (i)-x mea (i)|,i=1,2,...,N} (5)

[0126] In the formula, σ mxe Indicates the maximum deviation.

[0127] (3) Weighted average deviation

[0128] σ G_me =0.1σ me_pre +0.6σ me_fault +0.1σ me_post (6)

[0129] In the formula, σ G_me represents the weighted mean deviation, σ me_pre , σ me_fault , σ me_post They represent the average deviation before, during, and after a fault, respectively.

[0130] (4) Weighted maximum deviation

[0131] σ G_mxe =0.1σ mxe_pre +0.6σ mxe_fault +0.1σ mxe_post (6)

[0132] In the formula, σ G_mxe represents the weighted mean deviation, σ mxe_pre , σ mxe_fault , σ mxe_post They represent the average deviation before, during, and after a fault, respectively.

[0133] Embodiment 2 of the present invention provides a digital-analog hybrid simulation electromechanical transient modeling system for an energy storage system, including: an energy storage system fault ride-through test module, an energy storage system fault ride-through parameter identification module, and an energy storage system fault ride-through model error calculation module;

[0134] The energy storage system performance test module is used to test the energy storage system response characteristics under different grid strengths, different energy storage operating states, and different voltage drop / rise types and depths, including active power, reactive power, reactive current, etc.;

[0135] The energy storage system fault ride-through parameter identification module is used to identify the function expression parameters between the active current, reactive current and voltage, reactive power of the energy storage system during the fault voltage ride-through period;

[0136] The energy storage system fault ride-through model error calculation module is used to calculate the error between the energy storage system electromechanical transient model simulation results and the digital-analog hybrid simulation test results, and the comparison indicators include multiple electrical quantities such as voltage and power.

[0137] In order to verify the effectiveness of the method of the present invention, an energy storage system in an actual energy storage power station was selected to test and verify the scheme of the present invention.

[0138] The basic situation of the test system is as follows.

[0139] The rated capacity of the energy storage system is a 0.63MW wind turbine, the rated voltage at the machine end is 380V, and the rated frequency is 50Hz.

[0140] Using regression analysis, the fitting results of active current parameters are k1 = -0.0236, k2 =

[0141] 13.4858, the constant term is -11.3385; the fitting results of reactive current parameters are k1=0.0124, k2=0, and the constant term is -0.0082. That is, the expressions of the two are:

[0142] I p =-0.0236U t +13.4858I p0 -11.3385

[0143] I q =0.0124(U t_th -U t )+0*I q0 -0.0082

[0144] Where U t Indicates the output voltage amplitude of the energy storage system, U t_th Indicates the voltage threshold for entering low voltage ride-through, I p0 ,I q0 They represent the active current and reactive current before the voltage drops respectively.

[0145] In order to verify the effectiveness of the above method, the results of two low voltage ride-through are analyzed.

[0146] Set high power, discharge, and two-phase drop to 25%Un. The comparison between the electromechanical transient model simulation results and the test results is as follows Figure 2 The error calculation results are shown in Table 1.

[0147] Table 1. Calculation results of model error for high power, discharge, and two-phase drop to 25%Un

[0148]

[0149] Set high power, charge, and three-phase drop to 40%Un. The comparison between the electromechanical transient model simulation results and the test results is as follows Figure 3 The error calculation results are shown in Table 2.

[0150] Table 2. Calculation results of model errors for high power, charging, and three-phase drop to 40% Un

[0151]

[0152] It should be noted that due to the large fluctuations in the test model, some indicators exceeded the standards, indicating that the parameters of the energy storage system of the energy storage power station were not set reasonably and needed to be further adjusted to meet the requirements of relevant indicators.

[0153] According to the comparison results, it can be seen that the method of the present invention can establish an accurate electromechanical transient simulation model of a wind turbine, which verifies the superiority of the method proposed by the present invention.

[0154] The beneficial effect of the present invention is that, compared with the prior art,

[0155] 1. According to the operating characteristics of the energy storage system under different active power output levels, a set of energy storage system test conditions that meet the modeling requirements is constructed to provide sufficient and reliable test data for energy storage system modeling;

[0156] 2. To address the problem of black-box control strategy of energy storage systems, the voltage thresholds for energy storage systems to enter / exit low voltage ride-through and high voltage ride-through are identified by reducing the voltage drop step size;

[0157] 3. The topology structure, component parameters, controller type, and control program software version information of the energy storage system test model are consistent with those on site, making the established model more consistent with actual operating characteristics;

[0158] 4. Establish a parameter identification method and model evaluation index for the electromechanical transient simulation model of the energy storage system based on regression analysis to improve the model simulation accuracy while comprehensively reflecting the accuracy of the energy storage system model.

[0159] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents, and any modifications or equivalent replacements that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A hybrid digital-analog simulation electromechanical transient modeling method for energy storage system, characterized in that: include: Step A1, collect data on the actual energy storage power station, including the energy storage system controller model and software version number, primary circuit topology and component parameters, establish a hybrid digital simulation test model and a physical model of the energy storage system, and check the signal channel between the digital simulation test model and the physical model; Step A2, setting a fault test condition, performing a digital-analog hybrid simulation fault ride-through test on the energy storage system, recording the test results, and obtaining the electromechanical transient simulation model parameters of the energy storage system based on regression analysis; Step A3: construct an electromechanical transient simulation model of the energy storage system, compare the operation results of the electromechanical transient simulation model of the energy storage system and the digital-analog hybrid model under different working conditions, and calculate the error between the two.

2. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 1, characterized in that: The step A1 comprises: Step A1.1, based on the energy storage power station information, establish a digital-analog hybrid simulation test model of the same type of energy storage system, and check the signal channels between the digital simulation model and the physical device one by one; Step A1.2, check the signal channel of the digital-analog hybrid simulation test model. When the given value of the digital model and the measured value of the energy storage controller are inconsistent, modify the transformation ratio of the digital channel of the digital model.

3. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 2, characterized in that: In step A1.1, the energy storage system digital-analog hybrid simulation test model includes a primary circuit model module and an energy storage system control model module. The primary circuit model module includes a circuit model unit and an energy storage system body model unit; The energy storage system control model module includes controller hardware consistent with the field controller model and a control program version consistent with the field controller control program.

4. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 3, characterized in that: The circuit model unit includes a power grid model, a transformer model and output end inductor and capacitor elements, and the energy storage system body model unit includes a battery model, a converter model and a DC side model.

5. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 1, characterized in that: The step A2 comprises: Step A2.1, setting the electromechanical transient modeling test conditions of the energy storage system, including the grid strength, the output power level of the energy storage system, the output voltage change amplitude and duration, and completing the digital-analog hybrid simulation fault ride-through test of the energy storage system under different working conditions; Step A2.2, according to the hybrid digital-analog simulation fault ride-through test results of the energy storage system, a regression analysis method based on gradient descent is used to obtain the electromechanical transient simulation model parameters of the energy storage system; The gradient descent method is used to realize the multivariate regression of the parameters of the electromechanical transient model of the energy storage system, and multiple influencing factors X = [x1, x2, ..., x n ] and the dependent variable Y=[y1,y2,...,y m ], the weight relationship between them is: y k =ω0+ω1x1+ω2x2+...+ω n x n In the formula, ω i represents the weight of the i-th influencing factor; X=[x1,x2,...,x n ] represents the test results of the energy storage system digital-analog hybrid simulation, including the voltage in the steady-state period during the fault and the reactive power before the fault. Y = [y1, y2, ..., y m ] represents the active current amplitude and reactive current amplitude in the steady-state period during the fault period; The vector calculation expression of the multivariate linear regression model is: y k =ω T X In the formula, ω T represents the weight factor, ω=[ω0,ω1,ω2,...,ω n ]; The loss function of multivariate linear regression uses mean square error, which is expressed as: The condition for the loss function to be minimum is that the derivative of the loss function with respect to the weight factor is 0, and the calculation expression is: The normal equation of the multiple regression equation is obtained as follows: ω=(X T X) -1 X T Y The gradient descent method is used to solve the problem, and the loss function is used to derive the weight coefficient. The iterative calculation expression of the weight factor based on the gradient descent algorithm is: In the formula, η represents the learning rate; By iteratively updating the gradient and weight factor, the weight factor optimization result is finally obtained. The weight factor optimization result is the coefficient of the function expression of the active current and reactive current with the pre-fault current and the voltage amplitude during the fault period during the steady-state period of voltage change; by controlling the voltage drop amplitude step size, m <n。 6. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 5, characterized in that: The step A2.1 includes the following requirements for the output power and voltage change of the energy storage system: (1) Active power: The operating state includes charging state and discharging state, and the output power range includes 0.1P N ≤P≤0.3P N , 0.4P N ≤P≤0.7P N 、P>0.8P N Three types; (2) Voltage drop / increase setting: The voltage drop amplitude is in steps of 0.1 pu, and the voltage drop range is 0.1 pu-0.9 pu; the voltage increase amplitude is in steps of 0.05 pu, and the voltage increase range is 1.1 pu-1.3 pu; (3) Entering / Exiting Fault Voltage Ride-Threshold: The energy storage system enters / exits the low voltage ride-through voltage threshold test, with the voltage drop in steps of 0.01 pu and a voltage drop range of 0.8 pu-0.9 pu. The energy storage system enters / exits the high voltage ride-through voltage threshold test, with the voltage increase in steps of 0.01 pu and a voltage drop range of 1.1 pu-1.2 pu.

7. The method for electromechanical transient modeling of a digital-analog hybrid simulation of an energy storage system according to claim 6, characterized in that: The step A3 comprises: Step A3.1, calculating the electromechanical transient model error of the energy storage system, including the voltage, active power, reactive power, and reactive current of the output port of the energy storage system; Step A3.2, the error types include the following 4 types: (1) Average deviation In the formula, σ me represents the average deviation, the subscript sim represents the model simulation result, the subscript mea represents the digital-analog hybrid simulation test result, and N represents the number of test data; (2) Maximum deviation σ mxe =max{x sim (i)-x mea (i),i=1,2,...,N} In the formula, σ mxe represents the maximum deviation; (3) Weighted average deviation s G_me =0.1σ me_pre +0.6σ me_fault +0.1σ me_post In the formula, σ G_me represents the weighted mean deviation, σ me_pre , σ me_fault , σ me_post They represent the average deviation before, during, and after the fault, respectively; (4) Weighted maximum deviation s G_mxe =0.1σ mxe_pre +0.6σ mxe_fault +0.1σ mxe_post In the formula, σ G_mxe represents the weighted mean deviation, σ mxe_pre , σ mxe_fault , σ mxe_post They represent the average deviation before, during, and after the fault, respectively.

8. A digital-analog hybrid simulation electromechanical transient modeling system for an energy storage system, the system being the system used in the digital-analog hybrid simulation electromechanical transient modeling method for an energy storage system as claimed in any one of claims 1 to 7, characterized in that: include: Energy storage system performance test module, energy storage system fault ride-through parameter identification module, energy storage system fault ride-through model error calculation module; The energy storage system performance test module is used to test the energy storage system response characteristics under different grid strengths, different energy storage operating states, and different voltage drop / rise types and depths, including active power, reactive power, and reactive current; The energy storage system fault ride-through parameter identification module is used to identify the function expression parameters between the active current, reactive current and voltage and reactive power of the energy storage system during the fault voltage ride-through period; The energy storage system fault ride-through model error calculation module is used to calculate the error between the energy storage system electromechanical transient model simulation result and the digital-analog hybrid simulation test result, and the comparison indicators include voltage and power.

9. A terminal comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images