Droop coefficient optimization model construction method and droop coefficient online adjustment method

Abstract

This invention provides a method for constructing a droop coefficient optimization model and an online adjustment method for the droop coefficient, relating to the field of power electronics technology. The method for constructing the droop coefficient optimization model includes: calculating the operating point and corresponding droop coefficient under different operating conditions based on the parameter constraints of the power grid interconnection system; obtaining the impedance of the power grid interconnection system and calculating the corresponding weakest pole based on the impedance and droop coefficient; training an initial static optimization model and an initial stability judgment model using the impedance, operating point, droop coefficient, and weakest pole as a dataset to obtain a static optimization model for the droop coefficient and a stability judgment model; and constructing an initial impedance model and adjusting its parameters using the Kalman filter algorithm to obtain an impedance identification model. The beneficial effects of this invention are: expanding the applicability of the droop coefficient optimization model, and enabling online optimization and adjustment of the droop coefficient under various conditions, thereby achieving stable operation of the power grid interconnection system.

Description

Technical Field

[0001] This invention relates to the field of power electronics technology, and more specifically, to a method for constructing a droop coefficient optimization model and a method for online adjustment of the droop coefficient. Background Technology

[0002] With the development of new energy technologies, the proportion of new energy power electronic converters in power systems is gradually increasing. Under weak grid conditions, the output power of new energy generation is often limited, requiring the injection of reactive power to boost the grid connection voltage and ensure normal power transmission. However, to improve applicability, new energy converters often need to adapt to grids with a wide range of short-circuit ratios, necessitating the power electronic converter's ability to sense grid conditions. In some cases, grid conditions change over time. Existing technologies use methods that design control parameters under fixed operating conditions to control the normal operation of the grid system. However, grid parameters change over time during operation, leading to grid instability, which in turn causes instability in the converter and makes power transmission prone to anomalies. Summary of the Invention

[0003] The problem addressed by this invention is how to ensure the operational stability of a power grid interconnection system.

[0004] To address the aforementioned problems, this invention provides a method for constructing a droop coefficient optimization model and a method for online adjustment of the droop coefficient.

[0005] In a first aspect, the present invention provides a method for constructing a droop coefficient optimization model, wherein the droop coefficient optimization model includes a static droop coefficient optimization model and a stability judgment model, and the method for constructing the droop coefficient optimization model includes:

[0006] Based on the parameter limitations of the power grid interconnection system, operating points under different operating conditions are generated. The operating points include the output voltage, output current, and power of the power grid interconnection system during stable operation. The parameter limitations include transmission power limit, current limit, voltage limit, and modulation ratio limit.

[0007] Generate the corresponding droop coefficient based on the working point;

[0008] Obtain the impedance of the power grid interconnection system, and determine the corresponding weakest pole based on the impedance and the droop coefficient, wherein the impedance includes the inductance of the power grid interconnection system;

[0009] A preset initial static optimization model is trained using the operating point, the impedance, and the droop coefficient as a dataset to obtain the droop coefficient static optimization model. A preset initial stability judgment model is trained using the impedance, the operating point, the droop coefficient, and the weakest pole as a dataset to obtain the stability judgment model. The droop coefficient static optimization model is used to output the optimized droop coefficient and the range of values ​​for the optimized droop coefficient. The stability judgment model is used to output the weakest pole under different operating conditions. The range of values ​​for the optimized droop coefficient is the interval between the minimum and maximum droop coefficients determined according to the parameter constraints.

[0010] The method for constructing the droop coefficient optimization model also includes:

[0011] An initial impedance model is constructed based on the Kalman filter algorithm, and the parameters of the initial impedance model are adjusted to obtain the impedance identification model.

[0012] In this invention, an initial static optimization model is trained using the stable operating points of the interconnected power grid system under different operating conditions, the corresponding droop coefficients, and impedances, resulting in a droop coefficient static optimization model. An initial stability judgment model is trained using the stable operating points of the interconnected power grid system under different operating conditions, the corresponding droop coefficients, impedances, and the weakest pole representing the stability boundary of the interconnected power grid system, resulting in a stability judgment model. These two models constitute the droop coefficient optimization model. This model considers both the output power and stability of the interconnected power grid system, effectively expanding its applicability. Furthermore, when adjusting the droop coefficient, the droop coefficient that ensures both output power and stable operation can be obtained based on the real-time operating conditions of the interconnected power grid system. Adjusting the droop coefficient ensures the normal operation of the interconnected power grid system. Additionally, an impedance identification model is constructed and its parameters are tuned using the Kalman filter algorithm, effectively reducing the identification error and improving the accuracy of the impedance identification model, resulting in more reliable power grid impedance identification results.

[0013] Optionally, the grid interconnection system includes an inductive grid, an inverter, and a converter. One end of the inverter is connected to the converter, and the other end of the inverter is connected to the inductive grid. The connection point between the inverter and the inductive grid is the grid connection point. Generating operating points under different operating conditions based on the parameter limitations of the grid interconnection system includes:

[0014] min:-|U s -U n |

[0015]

[0016] U s,min ≤U s ≤Us,max

[0017] η min ≤η≤η max

[0018] P s (I sd U s ) = P o ,

[0019]

[0020] i cd =i sd i cq =i sq +B f U s

[0021]

[0022] P s =U s I sd ,

[0023] Among them, U s U represents the grid connection point voltage. n I represents the rated output voltage of the inverter. cd I represents the output current of the inverter on the d-axis in the dq coordinate system. cq I represents the output current of the inverter on the q-axis in the dq coordinate system. lim U represents the output current limiting of the converter. s，max U s，min Indicates the grid connection point voltage limit, P s I represents the active power output by the inverter to the inductive grid. sd P represents the grid connection point current on the d-axis in the dq coordinate system. o U is the limit of power transfer from the inverter to the inductive grid. g R represents the equivalent voltage of the power grid. g I represents the equivalent reactance of the power grid. sq X represents the grid-connected point current on the q-axis in the dq coordinate system. g η represents the equivalent resistance of the power grid, and η represents the modulation ratio. min Indicates the minimum modulation ratio limit, η max Indicates the maximum modulation ratio limit, i cd Indicates the magnitude of the d-axis component of the converter output current, i sd Indicates the magnitude of the d-axis component of the current transmitted to the power grid, i cq Indicates the magnitude of the q-axis component of the converter output current, i sqB represents the magnitude of the q-axis component of the current transmitted to the power grid. f U represents the susceptance of the converter filter capacitor. c X represents the inverter output voltage. f This represents the converter filter inductance.

[0024] Optionally, obtaining the impedance of the power grid interconnection system and determining the corresponding weakest pole based on the impedance and the droop coefficient includes:

[0025] Obtain the output impedance model of the power grid interconnection system;

[0026] Based on the output impedance model, the impedance of the power grid interconnection system is obtained;

[0027] Obtain the droop coefficient under different operating conditions, and determine the corresponding poles under different operating conditions based on the impedance and the droop coefficient.

[0028] Optionally, determining the corresponding weakest pole based on the impedance and the droop coefficient further includes:

[0029] Obtain a complex coordinate system for the poles, wherein the poles are represented in the complex coordinate system;

[0030] Determine the distance between each of the poles and the imaginary axis of the complex coordinate system of the poles;

[0031] The pole with the shortest distance is selected as the weakest pole.

[0032] Optionally, the initial impedance model includes discrete iterative equations and observation equations. The step of constructing the initial impedance model using the Kalman filter algorithm and adjusting the parameters of the initial impedance model to obtain the impedance identification model includes:

[0033] Based on the power grid circuit model, the circuit equations are obtained;

[0034] Based on the circuit equations, the discrete iterative equations are obtained;

[0035] Data observations are performed on the grid connection points to establish the observation equations;

[0036] The impedance identification model is obtained by adjusting and optimizing the parameters based on the observation equation and the discrete iterative equation.

[0037] Secondly, the present invention also provides a droop coefficient optimization model construction device, including a memory and a processor;

[0038] The memory is used to store computer programs;

[0039] The processor is configured to implement the droop coefficient optimization model construction method as described above when executing the computer program.

[0040] The droop coefficient optimization model construction method and the droop coefficient optimization model construction device described in this invention have the same advantages over the prior art, and will not be repeated here.

[0041] Thirdly, the present invention also provides a method for online adjustment of the sag coefficient, comprising:

[0042] The impedance is obtained using the impedance identification model established based on the above-mentioned droop coefficient optimization model construction method;

[0043] Obtain the parameter constraints of the power grid interconnection system, and use the parameter constraints and the impedance input to establish a droop coefficient optimization model based on the droop coefficient optimization model construction method described above, to obtain the optimized droop coefficient, the value range of the droop coefficient, and the weakest pole;

[0044] A stability index is set, and the droop coefficient in the power grid interconnection system is adjusted based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient.

[0045] In this invention, an impedance identification model is used to obtain accurate impedance in real time, enabling precise optimization of the droop coefficient. Based on the droop coefficient optimization model and the impedance, the optimized droop coefficient, its value range, and the corresponding weakest pole are obtained. The initial droop coefficient is adjusted according to stability indicators. This achieves adaptive power transmission under wide-range grid conditions while ensuring the stable operation of the grid interconnection system. Furthermore, the use of the droop coefficient optimization model effectively reduces the computational burden, lowers complexity, reduces usage costs, and improves practical applicability.

[0046] Optionally, adjusting the droop coefficient in the power grid interconnection system based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient includes:

[0047] Based on the real part of the weakest pole and the stability index, determine whether the power grid interconnection system meets the stability margin requirements;

[0048] If the real part of the weakest pole is greater than or equal to the stability index, then the power grid interconnection system does not meet the stability margin requirement.

[0049] If the real part of the weakest pole is less than the stability index, then the power grid interconnection system meets the stability margin requirement.

[0050] Optionally, adjusting the droop coefficient in the power grid interconnection system based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient includes:

[0051] If the power grid interconnection system meets the stability margin requirement, then the droop coefficient in the power grid interconnection system is adjusted according to the optimized droop coefficient.

[0052] If the power grid interconnection system does not meet the stability margin requirement, the droop coefficient is adjusted according to the range of change of the droop coefficient, and the process returns to the step of determining whether the power grid system meets the stability margin requirement until the power grid interconnection system meets the stability margin requirement.

[0053] Fourthly, the present invention also provides an online sag coefficient adjustment device, including a memory and a processor;

[0054] The memory is used to store computer programs;

[0055] The processor is configured to implement the online adjustment method for the droop coefficient as described above when executing the computer program.

[0056] The online sag coefficient adjustment device and the online sag coefficient adjustment method described in this invention have the same advantages over the prior art, and will not be repeated here. Attached Figure Description

[0057] Figure 1 This is a flowchart illustrating the method for constructing a droop coefficient optimization model according to an embodiment of the present invention. Figure 1 ;

[0058] Figure 2 This is a schematic diagram of the power grid interconnection system according to an embodiment of the present invention;

[0059] Figure 3 This is a flowchart illustrating the online adjustment method for the sag coefficient according to an embodiment of the present invention. Figure 1 ;

[0060] Figure 4 This is a schematic diagram illustrating the power grid impedance identification effect of the impedance identification model based on Kalman filtering according to an embodiment of the present invention.

[0061] Figure 5 This is a schematic diagram illustrating the effect of online adjustment of the sag coefficient according to an embodiment of the present invention;

[0062] Figure 6 This is a schematic diagram of the curved surface effect of the real part of the weakest pole under different inductance and droop coefficients in an embodiment of the present invention;

[0063] Figure 7 This is a schematic diagram illustrating the effect of online adjustment of the droop coefficient at the grid connection point voltage according to an embodiment of the present invention. Figure 1 ;

[0064] Figure 8 This is a schematic diagram illustrating the effect of online adjustment of grid connection point voltage without droop coefficient according to an embodiment of the present invention;

[0065] Figure 9 This is a schematic diagram illustrating the effect of online adjustment of grid connection point voltage without droop coefficient according to an embodiment of the present invention;

[0066] Figure 10 This is a schematic diagram illustrating the effect of online adjustment of the droop coefficient at the grid connection point voltage according to an embodiment of the present invention. Figure 2 ;

[0067] Figure 11 This is a flowchart illustrating the method for constructing a droop coefficient optimization model according to an embodiment of the present invention. Figure 2 ;

[0068] Figure 12 This is a flowchart illustrating the online adjustment method for the sag coefficient according to an embodiment of the present invention. Figure 2 . Detailed Implementation

[0069] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0070] In the description of the embodiments in this application, the term "some embodiments" means that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same implementation or instance. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0071] like Figure 1 and Figure 11 As shown, this embodiment of the invention provides a method for constructing a droop coefficient optimization model, the method comprising:

[0072] Step S1: Based on the parameter limitations of the power grid interconnection system, generate operating points under different operating conditions. The operating points include the output voltage, output current, and power of the power grid interconnection system during stable operation. The parameter limitations include transmission power limit, current limit, voltage limit, and modulation ratio limit.

[0073] Step S2: Generate the corresponding droop coefficient based on the working point;

[0074] Step S3: Obtain the impedance of the power grid interconnection system, and determine the corresponding weakest pole based on the impedance and the droop coefficient, wherein the impedance includes the inductance of the power grid interconnection system;

[0075] Step S4: Train a preset initial static optimization model using the operating point, the impedance, and the droop coefficient as the dataset to obtain the droop coefficient static optimization model. Train a preset initial stability judgment model using the impedance, the operating point, the droop coefficient, and the weakest pole as the dataset to obtain the stability judgment model. The droop coefficient static optimization model is used to output the optimized droop coefficient and the range of the optimized droop coefficient. The stability judgment model is used to output the weakest pole under different operating conditions. The range of the optimized droop coefficient is the interval between the minimum and maximum droop coefficients determined according to the parameter constraints.

[0076] Step S5: Construct an initial impedance model based on the Kalman filter algorithm, and adjust the parameters of the initial impedance model to obtain the impedance identification model.

[0077] Specifically, since the output voltage and current of a power grid interconnection system are actually limited, a reasonable configuration of the output voltage and current is necessary to achieve stable output power. The operating point of a power grid interconnection system is the state of its steady-state voltage and current, i.e., the state of stable output power. Therefore, the operating point can represent the output parameters of the power grid interconnection system at its optimal stability, including output voltage, output current, and output power. A dq coordinate system is established for the power grid system; the component of the output current on the q-axis corresponds to reactive power, and correspondingly, the component of the output current on the d-axis corresponds to active power. The droop coefficient obtained from the operating point is also the optimal droop coefficient. By generating the operating point based on the parameter constraints of the power grid interconnection system and determining the droop coefficient, the range of droop coefficient values ​​under different operating conditions can be obtained for the entire power grid interconnection system during stable operation. Each operating condition has a corresponding optimal droop coefficient. For example, based on voltage constraints, two corresponding extreme operating points are obtained, and the corresponding minimum and maximum droop coefficients are obtained from these two extreme operating points. The interval between these two values ​​represents the range of droop coefficient values. The grid connection point is the interconnection point in the power grid interconnection system. In droop control, the voltage at the grid connection point and the output reactive power must satisfy the following:

[0078] Q = K v (U n -U s ),

[0079] Among them, K v U represents the droop coefficient. n Represents the reference voltage; U s Q represents the grid connection point voltage, and Q represents the output reactive power.

[0080] Based on the above derivation, the droop coefficient is expressed as:

[0081]

[0082] Furthermore, for a specific operating point, all poles of the power grid interconnection system can be calculated, with the number of poles corresponding to the order of the interconnection system. The poles are the basis for judging the stability of the operating point; typically, the weakest pole is the closest to instability among all poles. Therefore, the weakest pole is determined based on the grid impedance, the operating point, and the corresponding droop coefficient, and this is used as a judgment of the stability of the power grid interconnection system. In this embodiment, a static optimization model for the droop coefficient is trained using inductance, operating points under different operating conditions, and the corresponding droop coefficients as datasets. A stability judgment model is trained using inductance, droop coefficients, and the corresponding weakest pole as datasets. The static optimization model for the droop coefficient takes inductance and power at the operating point as inputs and outputs the optimized droop coefficient and its value range. The stability judgment model takes inductance, power at the operating point, and the droop coefficient as inputs and outputs the weakest pole. The stability judgment model and the droop coefficient static optimization model together constitute the droop coefficient optimization model. This droop coefficient optimization model can determine the optimal droop coefficient, the range of values ​​for the droop coefficient, and the weakest pole for different operating conditions of the power grid interconnection system. It can preset stability indicators, judge the stability of the power grid interconnection system when operating with the given droop coefficient based on the stability indicators and the weakest pole, and then adjust the droop coefficient in the power grid interconnection system online based on the stability judgment results.

[0083] In this embodiment, an initial static optimization model is trained using the stable operating points of the interconnected power grid system under different operating conditions, the corresponding droop coefficients, and impedances, resulting in a droop coefficient static optimization model. An initial stability judgment model is also trained using the stable operating points of the interconnected power grid system under different operating conditions, the corresponding droop coefficients, impedances, and the weakest pole representing the stability boundary of the interconnected power grid system, resulting in a stability judgment model. These two models constitute the droop coefficient optimization model, which considers both the output power and stability of the interconnected power grid system while ensuring its output power. This effectively expands the applicability of the droop coefficient optimization model. Furthermore, when adjusting the droop coefficient, the droop coefficient that ensures both output power and stable operation of the interconnected power grid system can be obtained based on the real-time operating conditions of the interconnected power grid system. Adjusting the droop coefficient ensures the normal operation of the interconnected power grid system. Additionally, an impedance identification model is constructed and its parameters are tuned using the Kalman filter algorithm, effectively reducing the identification error of the impedance identification model, improving its accuracy, and obtaining more reliable power grid impedance identification results.

[0084] Optionally, the grid interconnection system includes an inductive grid, an inverter, and a converter. One end of the inverter is connected to the converter, and the other end of the inverter is connected to the inductive grid. The connection point between the inverter and the inductive grid is the grid connection point. Generating operating points under different operating conditions based on the parameter limitations of the grid interconnection system includes:

[0085] min:-|U s -U n |

[0086]

[0087] U s,min ≤U s ≤U s,max

[0088] η min ≤η≤η max

[0089] P s (I sd U s ) = P o ,

[0090]

[0091] i cd =i sd i cq =i sq +B f U s

[0092]

[0093] P s =U s I sd ,

[0094] Among them, U s U represents the grid connection point voltage. n I represents the rated output voltage of the inverter. cd I represents the output current of the inverter on the d-axis in the dq coordinate system. cq I represents the output current of the inverter on the q-axis in the dq coordinate system. lim U represents the output current limiting of the converter. s，max U s，min Indicates the grid connection point voltage limit, P s I represents the active power output by the inverter to the inductive grid. sd P represents the grid connection point current on the d-axis in the dq coordinate system. o U is the limit of power transfer from the inverter to the inductive grid.g R represents the equivalent voltage of the power grid. g I represents the equivalent reactance of the power grid. sq X represents the grid-connected point current on the q-axis in the dq coordinate system. g η represents the equivalent resistance of the power grid, and η represents the modulation ratio. min Indicates the minimum modulation ratio limit, η max Indicates the maximum modulation ratio limit, i cd Indicates the magnitude of the d-axis component of the converter output current, i sd Indicates the magnitude of the d-axis component of the current transmitted to the power grid, i cq Indicates the magnitude of the q-axis component of the converter output current, i sq B represents the magnitude of the q-axis component of the current transmitted to the power grid. f U represents the susceptance of the converter filter capacitor. c X represents the inverter output voltage. f This represents the converter filter inductance.

[0095] like Figure 2 The diagram shown is a schematic diagram of the grid interconnection system according to an embodiment of the present invention. The converter is a new energy converter, and the inverter adopts a passive filter structure. The DC side of the converter is equivalent to a DC voltage source. The control structure adopts a traditional grid-following control structure, including an active power outer loop and a reactive power-droop control outer loop.

[0096] Furthermore, before constructing the aforementioned optimization model, a dq coordinate system needs to be established using a phase-locked loop (PLL) to ensure that the d-axis component of the established dq coordinate system coincides with the grid-connected point vector voltage, thus achieving d-axis voltage orientation. In actual systems, if the equivalent inductance of the grid changes abruptly, the current may exceed the limit value (and due to inductance identification errors, the current may also increase). Therefore, a relatively conservative value needs to be chosen when setting the current limiting condition, or a corresponding limiting element needs to be added to the current control loop. The above formula can be used to calculate the two extreme operating points corresponding to the grid-connected point voltage under given impedance and inverter output power values, providing a parameter variation range for subsequent stability adjustment steps.

[0097] In this embodiment, by determining the operating point based on parameter constraints, the operating state of the power grid interconnection system during stable operation can be obtained. The droop coefficient is then determined based on this operating state, providing a stable and accurate data foundation for the subsequent training of the droop coefficient optimization model. This allows for the construction of a droop coefficient optimization model that considers stability. Simultaneously, considering parameter constraints, the range of parameter variation can be obtained when using the droop coefficient optimization model for subsequent droop coefficient optimization, enabling accurate adjustment under a wider range of power grid conditions and ensuring the normal operation of the power grid.

[0098] Optionally, obtaining the impedance of the power grid interconnection system and determining the corresponding weakest pole based on the impedance and the droop coefficient includes:

[0099] Obtain the output impedance model of the power grid interconnection system;

[0100] Based on the output impedance model, the impedance of the power grid interconnection system is obtained;

[0101] Obtain the droop coefficient under different operating conditions, and determine the corresponding poles under different operating conditions based on the impedance and the droop coefficient.

[0102] Optionally, determining the corresponding weakest pole based on the impedance and the droop coefficient further includes:

[0103] Obtain a complex coordinate system for the poles, wherein the poles are represented in the complex coordinate system;

[0104] Determine the distance between each of the poles and the imaginary axis of the complex coordinate system of the poles;

[0105] The pole with the shortest distance is selected as the weakest pole.

[0106] Specifically, in offline mode, an output impedance model of the grid interconnection system is constructed. Based on the output impedance model, the Thevenin equivalent is applied to the grid interconnection system at the grid connection point to obtain the relationship between voltage disturbance and current disturbance:

[0107]

[0108] Among them, v d V represents the grid-connected point disturbance voltage along the d-axis in the dq coordinate system. q i represents the grid-connected point disturbance voltage along the q-axis in the dq coordinate system. d i represents the grid-connected point disturbance current along the d-axis in the dq coordinate system. q Z represents the grid-connected point disturbance current along the q-axis in the dq coordinate system. dq This is the impedance matrix obtained based on the output impedance model.

[0109] The poles of a power grid interconnection system are represented as follows:

[0110]

[0111] Among them, Z g Z is the impedance matrix of the inductive power grid. c Let I be the converter impedance matrix, I be the identity matrix, and det denotes the determination of the matrix determinant.

[0112] Because Z in the above formula g Z cSince s is a matrix in the s-domain, expanding the above equation yields a univariate polynomial equation in s, the degree of which is determined by the order of the system, and its form is shown below:

[0113] a n s n +a n-1 s n-1 +a n-2 s n-2 …+a1s+a0=0,

[0114] Among them, a n Let represent the coefficients of the nth polynomial, and s represent the poles of the interconnected power grid system. The number of poles is equal to the order of the system.

[0115] Specifically, poles are represented in a complex coordinate system, where the vertical axis represents the imaginary axis and the horizontal axis represents the real axis. The stability of poles is typically determined based on the imaginary axis. If all poles lie in the right half-plane of the complex coordinate system, it indicates that the operating point corresponding to that pole is stable, meaning the power grid interconnection system is operating stably under the current parameters. The distances of each pole to the imaginary axis are obtained and compared, and the pole with the smallest distance from the imaginary axis is selected as the weakest pole. The weakest pole is the closest to being unstable among the poles.

[0116] In this embodiment, an output impedance model is established to obtain the output impedance of the power grid interconnection system under different operating conditions, thereby determining the corresponding poles, obtaining a large amount of accurate pole data, and using the distance between the poles and the imaginary axis to obtain the weakest pole as the basis for judging system stability. A droop coefficient stability judgment model is trained to obtain the weakest pole under different droop coefficients in the subsequent model use process, so as to judge the stability of the power grid interconnection system and ensure the stability of the power grid interconnection system.

[0117] Optionally, the initial impedance model includes discrete iterative equations and observation equations. The step of constructing the initial impedance model using the Kalman filter algorithm and adjusting the parameters of the initial impedance model to obtain the impedance identification model includes:

[0118] Based on the power grid circuit model, the circuit equations are obtained;

[0119] Based on the circuit equations, the discrete iterative equations are obtained;

[0120] Data observations are performed on the grid connection points to establish the observation equations;

[0121] The impedance identification model is obtained by adjusting and optimizing the parameters based on the observation equation and the discrete iterative equation.

[0122] Specifically, assuming the ideal grid voltage remains constant, by changing the state of the grid connection point and subtracting the operating states of the grid connection point before and after the change, we can obtain the circuit equation:

[0123]

[0124] Where, Δi d Δi represents the change in grid-connected current along the d-axis in the dq coordinate system. q Δu represents the change in grid-connected current along the q-axis in the dq coordinate system. pccd Δu represents the change in grid-connected voltage along the d-axis in the dq coordinate system. pccq ω represents the change in grid connection point voltage along the q-axis in the dq coordinate system, R represents the grid angular frequency, and L represents the grid equivalent resistance.

[0125] Based on the circuit equations above, the discrete iterative equations for the Kalman filter are obtained:

[0126]

[0127] Where l is the reciprocal of the equivalent inductance L of the power grid, T is the discrete interval time, q is the corresponding model error, and K represents the value of the quantity corresponding to the Kth step of the algorithm execution.

[0128] By observing the voltage and current at the grid connection point, the observation equations are obtained:

[0129]

[0130] In this equation, the left side represents the observed data, the first term on the right side represents the state variable, and the second term on the right side represents the observation error of the corresponding number, i.e., r is the observation error.

[0131] This invention provides a device for constructing a droop coefficient optimization model, including a memory and a processor;

[0132] The memory is used to store computer programs;

[0133] The processor is configured to implement the droop coefficient optimization model construction method as described above when executing the computer program.

[0134] The droop coefficient optimization model construction method and the droop coefficient optimization model construction device described in this invention have the same advantages over the prior art, and will not be repeated here.

[0135] like Figure 3 and Figure 12 As shown, this embodiment of the invention further provides a method for online adjustment of the sag coefficient, including:

[0136] Step A1: Obtain the impedance using the impedance identification model established based on the above-mentioned droop coefficient optimization model construction method;

[0137] Step A2: Obtain the parameter constraints of the power grid interconnection system, input the parameter constraints and the impedance into the droop coefficient optimization model established by the above droop coefficient optimization model construction method, and obtain the optimized droop coefficient, the value range of the droop coefficient and the weakest pole;

[0138] Step A3: Set a stability index, and adjust the droop coefficient in the power grid interconnection system according to the weakest pole, the stability index, the optimized droop coefficient, and the value range of the droop coefficient.

[0139] Specifically, the impedance of the interconnected power grid system is acquired in real time using an impedance identification model. This impedance and parameter constraints are then input into the constructed droop coefficient optimization model. The model outputs the optimal droop coefficient, its range, and the weakest pole under the current operating state of the interconnected power grid system. The weakest pole serves as the criterion for determining the stability of the interconnected power grid system. A stability index is set and compared with the output weakest pole to assess the system's stability. The droop coefficient is then adjusted based on the optimal droop coefficient and its range.

[0140] In this embodiment, an impedance identification model is used to obtain accurate impedance in real time, enabling precise optimization of the droop coefficient. Based on the droop coefficient optimization model and the impedance, the optimized droop coefficient, its value range, and the corresponding weakest pole are obtained. The initial droop coefficient is adjusted according to stability indicators. While ensuring the stable operation of the power grid interconnection system, adaptive power transmission under wide-range power grid conditions is achieved. Furthermore, the use of a droop coefficient optimization model trained based on artificial neural networks effectively reduces the computational burden, lowers complexity, reduces usage costs, and improves practical applicability.

[0141] It should be noted that this embodiment applies to situations where the grid connection strength of the inverter (converter) is uncertain when connected to an inductive grid, and normal active power transmission is required under a wide range of grid conditions. It is applicable when the grid connection conditions of the inverter (new energy converter) may change, and it is required to ensure normal power transmission, system grid connection stability, and that the grid connection point voltage meets the normal operating range within a large range of changes.

[0142] Optionally, adjusting the droop coefficient in the power grid interconnection system based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient includes:

[0143] Step A31: Determine whether the power grid interconnection system meets the stability margin requirements based on the real part of the weakest pole and the stability index;

[0144] Step A32: If the real part of the weakest pole is greater than or equal to the stability index, then the power grid interconnection system does not meet the stability margin requirement.

[0145] Step A33: If the real part of the weakest pole is less than the stability index, then the power grid interconnection system meets the stability margin requirement.

[0146] Step A34: If the power grid interconnection system meets the stability margin requirement, then adjust the droop coefficient in the power grid interconnection system according to the optimized droop coefficient;

[0147] Step A35: If the power grid interconnection system does not meet the stability margin requirement, adjust the droop coefficient according to the range of change of the droop coefficient, and return to the step of determining whether the power grid system meets the stability margin until the power grid interconnection system meets the stability margin requirement.

[0148] In this embodiment, the weakest pole and stability index (e.g., the real part of the pole is less than a preset value) are used to determine whether the power grid interconnection system has sufficient stability margin when running with the optimized droop coefficient. This ensures the accuracy of the droop coefficient optimization and avoids optimization errors that could lead to instability in the power grid interconnection system. By setting a stability margin, the power grid interconnection system is prevented from operating unstablely or at the brink of stability during actual operation, thus ensuring its stability. When the power grid interconnection system becomes unstable, the initial droop coefficient is adjusted according to its value range, and the stability is further assessed to ensure stable operation of the power grid interconnection system.

[0149] The present invention also provides an online sag coefficient adjustment device, including a memory and a processor;

[0150] The memory is used to store computer programs;

[0151] The processor is configured to implement the online adjustment method for the droop coefficient as described above when executing the computer program.

[0152] The online sag coefficient adjustment method and the online sag coefficient adjustment device described in this invention have the same advantages over the prior art, and will not be repeated here.

[0153] In a specific embodiment:

[0154] The structure diagram of the power grid interconnection system is as follows: Figure 2 As shown, the inverter adopts a passive filter structure. The DC side of the converter is equivalent to a DC voltage source. The control structure adopts a traditional grid-fed control structure, including an active power outer loop and a reactive power-droop control outer loop. The filter inductor L... f The filter capacitor C is 80uH. fIt is 160uF, DC voltage U dc The rated AC voltage is 1500V (i.e., the inverter output voltage U). c The voltage is 800V, and the rated power (i.e., the inverter output current I) is... c With inverter output voltage U c The product of ( ) is 230kW, and the droop control reference voltage is taken as 1.2 times the rated voltage (to avoid a negative droop coefficient). For better demonstration, an average value model is used in the simulation. Wherein, U s Indicates the grid connection point voltage, I s I represents the grid connection point current. Lf L represents the filter inductor current. g This represents the inductance of the power grid.

[0155] like Figure 4 As shown, a grid impedance identification technique based on extended Kalman filtering is employed, where L represents inductance and t represents time. Small reactive power pulses of 100ms in length are injected with a period of 1s, simultaneously initiating grid impedance estimation. In the simulation, the initial grid inductance is set to 5mH, increasing to 6mH at 2.3s and then to 6.5mH at 3.3s. The extended Kalman filter initiates its first estimation at 2s, ultimately obtaining an estimated value of 4.705mH; a second estimation is initiated at 3s, and after 100ms, an estimated value of 5.874mH is obtained; and estimation is initiated again at 4s, ultimately obtaining an estimated value of 6.344mH. It can be seen that the algorithm roughly tracks the changes in grid inductance.

[0156] The grid connection point voltage range is set to 0.95 pu to 1.1 pu, and the maximum current is 1.2 pu. This is achieved under different grid inductance (L... g Under the conditions of output power (P), the optimized droop coefficient K is obtained using the droop coefficient optimization model. v like Figure 5 As shown. It can be seen that L g Increasing both P and P will lead to an increase in the optimized droop coefficient, which is consistent with the results obtained from theoretical analysis.

[0157] like Figure 6 As shown, near-instability control parameters were selected for discussion. A pair of poles (the weakest pole, determined by whether the real part is greater than 0) in the grid interconnection system that may become unstable under parameter changes were fitted. The corresponding pole locations were determined under different output active power and droop coefficients, generating a dataset. The droop coefficient optimization model was then used to fit this dataset. With the output active power fixed at the rated power, different grid short-circuit ratios were plotted (in this embodiment, the grid inductance L). g (short-circuit ratio) and droop factor K vThe surface of the weakest pole real part shows that L g Increase, K v Increasing the size will make the system more unstable.

[0158] Using the online adjustment method for the droop coefficient in this embodiment, the droop coefficient and the waveform of the grid connection point voltage amplitude are as follows: Figure 7 As shown, Figure 7 (a) is a waveform diagram showing the dynamic adjustment of the droop coefficient based on the grid impedance identification results. After adjustment, the grid connection point voltage can be maintained within the range of 0.95 pu to 1.1 pu. Figure 7 (b) ensures the converter operates normally, with output power as... Figure 7 As shown in (c); without using the online adjustment method 1 for the droop coefficient, after increasing to 6.5mH in 3.3s, the short-circuit ratio and grid connection point voltage are too low, such as... Figure 8 (a) Its dynamic performance is too poor, resulting in instability, and the output power gradually diverges, such as Figure 8 (b); Under the condition where the droop coefficient is not adjusted online, method 2 increases the grid inductance to 6.2 mH in 3.3 s, and the voltage amplitude at the grid connection point further decreases to 0.9438 pu, which is less than the allowable range of 0.95 pu. Figure 9 (a) Although power can still be transmitted, such as Figure 9 (b) However, the system may experience problems due to voltage exceeding limits. Simulations illustrate the necessity of online adjustment of the droop coefficient under dynamic grid conditions.

[0159] like Figure 10 As shown, before 6.2s, the online adjustment method of the droop coefficient in this embodiment was not used in the simulation. After the grid short-circuit ratio changed to 1.3585 at 5.5s, the system began to oscillate and gradually developed. After applying the droop coefficient given by the droop coefficient optimization model in this embodiment at 6.2s, the voltage amplitude at the grid connection point of the system returned to the feasible range, and the system returned to a stable state. This demonstrates that the method can improve the stability of the system under weak grid conditions.

[0160] While the disclosure is as stated above, its scope of protection is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of this disclosure, and all such changes and modifications will fall within the protection scope of this invention.

Claims

1. A method for constructing a droop coefficient optimization model, characterized in that, The droop coefficient optimization model includes a static droop coefficient optimization model and a stability judgment model. The method for constructing the droop coefficient optimization model includes: Based on the parameter limitations of the power grid interconnection system, operating points under different operating conditions are generated. The operating points include the output voltage, output current, and power of the power grid interconnection system during stable operation. The parameter limitations include transmission power limit, current limit, voltage limit, and modulation ratio limit. Generate the corresponding droop coefficient based on the working point; Obtaining the impedance of the power grid interconnection system, and determining the corresponding weakest pole based on the impedance and the droop coefficient, includes: Obtain the output impedance model of the power grid interconnection system; Based on the output impedance model, the impedance of the power grid interconnection system is obtained; Obtain the droop coefficient under different operating conditions, and determine the corresponding poles under different operating conditions based on the impedance and the droop coefficient; Obtain a complex coordinate system for the poles, wherein the poles are represented in the complex coordinate system; Determine the distance between each of the poles and the imaginary axis of the complex coordinate system of the poles; Compare the distances and select the pole with the shortest distance as the weakest pole; Wherein, the impedance includes the inductance of the power grid interconnection system; A preset initial static optimization model is trained using the operating point, the impedance, and the droop coefficient as a dataset to obtain the droop coefficient static optimization model. A preset initial stability judgment model is trained using the impedance, the operating point, the droop coefficient, and the weakest pole as a dataset to obtain the stability judgment model. The droop coefficient static optimization model is used to output the optimized droop coefficient and the range of values ​​for the optimized droop coefficient. The stability judgment model is used to output the weakest pole under different operating conditions. The range of values ​​for the optimized droop coefficient is the interval between the minimum and maximum droop coefficients determined according to the parameter constraints. The method for constructing the droop coefficient optimization model also includes: An initial impedance model is constructed based on the Kalman filter algorithm, and the parameters of the initial impedance model are adjusted to obtain the impedance identification model.

2. The method for constructing a droop coefficient optimization model according to claim 1, characterized in that, The grid interconnection system includes an inductive grid, an inverter, and a converter. One end of the inverter is connected to the converter, and the other end of the inverter is connected to the inductive grid. The connection point between the inverter and the inductive grid is the grid connection point. The process of generating operating points under different operating conditions based on the parameter limitations of the power grid interconnection system includes: ， ， in, Indicates the voltage at the grid connection point. I represents the rated output voltage of the inverter. cd I represents the output current of the inverter on the d-axis in the dq coordinate system. cq This represents the inverter's output current along the q-axis in the dq coordinate system. Indicates the output current limiting of the converter. , Indicates the voltage limit at the grid connection point. This represents the active power output by the inverter to the inductive grid. This represents the grid connection point current along the d-axis in the dq coordinate system. This represents the power transfer limit from the inverter to the inductive grid. Indicates the equivalent voltage of the power grid. Represents the equivalent reactance of the power grid. This represents the grid connection point current on the q-axis in the dq coordinate system. η represents the equivalent resistance of the power grid, and η represents the modulation ratio. min Indicates the minimum modulation ratio limit, η max Indicates the maximum modulation ratio limit, i cd Indicates the magnitude of the d-axis component of the converter output current, i sd Indicates the magnitude of the d-axis component of the current transmitted to the power grid, i cq Indicates the magnitude of the q-axis component of the converter output current, i sq This indicates the magnitude of the q-axis component of the current transmitted to the power grid. This indicates the susceptance of the converter filter capacitor. Indicates the inverter output voltage. This represents the converter filter inductance.

3. The method for constructing a droop coefficient optimization model according to claim 2, characterized in that, The initial impedance model includes discrete iterative equations and observation equations. The initial impedance model is constructed using the Kalman filter algorithm, and parameters are adjusted to obtain the impedance identification model, which includes: Based on the power grid circuit model, the circuit equations are obtained; Based on the circuit equations, the discrete iterative equations are obtained; Data observations are performed on the grid connection points to establish the observation equations; The impedance identification model is obtained by adjusting and optimizing the parameters based on the observation equation and the discrete iterative equation.

4. A device for constructing a droop coefficient optimization model, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the droop coefficient optimization model construction method as described in any one of claims 1 to 3 when executing the computer program.

5. A method for online adjustment of the sag coefficient, characterized in that, include: The impedance is obtained by using the impedance identification model established by the droop coefficient optimization model construction method according to any one of claims 1 to 3; Obtain the parameter constraints of the power grid interconnection system, and use the parameter constraints and the impedance input to establish a droop coefficient optimization model according to any one of claims 1 to 3, to obtain the optimized droop coefficient, the value range of the droop coefficient, and the weakest pole. A stability index is set, and the droop coefficient in the power grid interconnection system is adjusted based on the weakest pole, the stability index, the optimized droop coefficient, and the value range of the droop coefficient.

6. The method for online adjustment of the sag coefficient according to claim 5, characterized in that, The adjustment of the droop coefficient in the power grid interconnection system based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient includes: Based on the real part of the weakest pole and the stability index, determine whether the power grid interconnection system meets the stability margin requirements; If the real part of the weakest pole is greater than or equal to the stability index, then the power grid interconnection system does not meet the stability margin requirement. If the real part of the weakest pole is less than the stability index, then the power grid interconnection system meets the stability margin requirement.

7. The method for online adjustment of the sag coefficient according to claim 6, characterized in that, The adjustment of the droop coefficient in the power grid interconnection system based on the weakest pole, the stability index, the optimized droop coefficient, and the range of values ​​for the droop coefficient includes: If the power grid interconnection system meets the stability margin requirement, then the droop coefficient in the power grid interconnection system is adjusted according to the optimized droop coefficient. If the power grid interconnection system does not meet the stability margin requirement, the droop coefficient is adjusted according to the range of change of the droop coefficient, and the process returns to the step of determining whether the power grid system meets the stability margin requirement, until the power grid interconnection system meets the stability margin requirement.

8. A device for online adjustment of sag coefficient, characterized in that, Including memory and processor; The memory is used to store computer programs; The processor is configured to implement the online adjustment method for the droop coefficient as described in any one of claims 5 to 7 when executing the computer program.

Images