Inverter parameter optimization method based on passivity theory and particle swarm optimization

Abstract

The invention discloses an inverter parameter optimization method based on a passivity theory and a particle swarm algorithm, and relates to the technical field of grid-connected inverter control system parameter optimization. Comprising the steps of constructing a full-band system model of the inverter, defining evaluation indexes by applying an expansibility passive theory, constructing an optimization objective function, executing a particle swarm optimization algorithm and outputting optimal parameters. According to the method, a rotation matrix in an extended passivity theory is introduced, so that a negative value failure problem of a traditional passivity index in a low-frequency band is converted into a positive value problem which can be optimally solved; by combining an extremely weak power grid model and a particle swarm global optimization algorithm, efficient and automatic setting of control parameters of the inverter is realized, so that the system is remarkably enhanced in a full frequency band, and the adaptability and robustness of the inverter under a weak power grid condition are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of grid-connected inverter control system parameter optimization technology, and in particular to an inverter parameter optimization method based on passive theory and particle swarm optimization algorithm. Background Technology

[0002] As the proportion of renewable energy in the power system continues to rise, the power grid gradually exhibits characteristics of low inertia and weak damping, and its intensity fluctuation range expands significantly, which can easily lead to wideband oscillations between grid-connected inverters and weak power grids.

[0003] To analyze and suppress such oscillations, frequency-domain-based passive theory has been widely adopted. Traditional input feedforward passivity (IFP) theory requires that the real part of the inverter's input admittance (also known as the passive index) be greater than zero across the entire frequency band to ensure stable operation of the system in any passive grid environment.

[0004] However, existing parameter tuning methods based on traditional passivity theory have several obvious technical limitations:

[0005] First, due to the influence of synchronization mechanisms such as phase-locked loops (PLLs) or power synchronous control (PSCs), grid-connected inverters exhibit inherent non-passive characteristics in the low-frequency range (generally near and below the base frequency), meaning their conventional passive index is always negative. This makes it impossible to directly establish an optimization function aimed at "maximizing the conventional passive index," because the optimization algorithm lacks effective gradient guidance in the negative range and is difficult to converge to the true stable region.

[0006] Second, in current engineering practice, attempts are often made to reduce the frequency range covered by the negative passive index by adjusting parameters. However, the latest research shows that simply reducing the negative value region cannot ensure system stability. On the contrary, it may induce new low-frequency oscillation modes due to over-adjustment of parameters such as active damping gain.

[0007] Third, there are complex nonlinear coupling relationships among the control parameters of the inverter (such as proportional coefficient, integral coefficient, feedforward gain, etc.). Traditional manual trial-and-error methods or methods based on a single frequency point index are difficult to find the global optimal solution that simultaneously takes into account low-frequency synchronization stability and high-frequency harmonic stability in the multi-parameter space.

[0008] Therefore, an inverter parameter optimization method based on passive theory and particle swarm optimization algorithm is designed to overcome the failure of traditional low-frequency indicators, quantify the stability margin across the entire frequency band, and automatically find the globally optimal control parameters. Summary of the Invention

[0009] The technical problem this invention aims to solve is to provide an inverter parameter optimization method based on passive property theory and particle swarm optimization (PSO). Its core innovation lies in transforming the system admittance using a rotation matrix, constructing the rotating passive property index from extended passive property theory as a full-frequency stability margin function, and using this as the fitness function for the PSO algorithm. Furthermore, leveraging the powerful global optimization capability of PSO, automated tuning of control parameters is achieved. This significantly improves the low-frequency synchronization stability of the inverter under weak grid conditions while ensuring high-frequency harmonic stability, thus solving the problem that traditional passive property indices cannot be directly used for parameter optimization in the low-frequency band.

[0010] The solution adopted by this invention to solve its technical problem is as follows:

[0011] An inverter parameter optimization method based on passivity theory and particle swarm optimization includes the following steps:

[0012] Step S1: Construct a full-band system model of the inverter.

[0013] Obtain the circuit topology parameters and initial control parameters of the inverter, and establish a second-order small-signal admittance matrix model Y(s) of the inverter in the dq coordinate system. The admittance matrix model Y(s) includes a phase-locked loop, an inner current loop, an outer voltage loop, and filter dynamics.

[0014] Step S2: Define evaluation metrics using the extended passive theory.

[0015] Set the boundary frequency The entire frequency band is divided into low-frequency and high-frequency bands. For the low-frequency band, a rotation matrix R from extended passive theory is introduced to perform a rotation transformation on the low-frequency admittance matrix. The eigenvalues ​​of the rotated admittance matrix are then calculated as stability indicators for the low-frequency band. For high-frequency bands, conventional passive performance indicators are used as stability indicators for high-frequency bands. Based on the pre-defined extreme grid impedance model, low-frequency margins are constructed respectively. and high-frequency margin ;

[0016] Step S3: Construct the optimization objective function.

[0017] Define the objective function For low-frequency margin With high-frequency margin The minimum value in the objective function To find the global minimum of the stability margin across the entire frequency band, and combining it with the rotation matrix R in step S2, the objective function is... The result is transformed into quantifiable positive values, and the optimization objective is to find the control parameter vector. ,make Maximize and exceed 0 within the given parameter range;

[0018] Step S4: Execute the particle swarm optimization algorithm.

[0019] Set the population size, maximum number of iterations, learning factor, and inertia weight of the particle swarm optimization algorithm. Randomly generate an initial particle population containing multiple sets of control parameters. In each iteration, calculate the objective function value of each particle under the corresponding control parameters as the fitness. Update the individual historical best position and the global historical best position. Use a dynamic inertia weight strategy to update the velocity and position of the particles and perform boundary constraint processing on the updated parameters.

[0020] Step S5: Output the optimal parameters.

[0021] Repeat step S4 until the fitness value meets the convergence condition, and output the combination of control parameters corresponding to the global optimal position as the final tuning parameters of the inverter.

[0022] As a preferred embodiment of the present invention

[0023] The low-frequency stability index in step S2 The calculation formula is:

[0024] ;

[0025] In the formula, Represents different parameters and different frequencies The extended passivity index below; Represents the Hermitian transformation of the rotation matrix R; It is composed of different parameters and different frequencies The complex function formed by these components represents the converter admittance values ​​under different parameters and frequencies, and is a second-order matrix. Represents the converter admittance matrix The Hermitian transform.

[0026] As a preferred embodiment of the present invention

[0027] In step S4, a linear decreasing strategy is used to dynamically adjust the inertia weight. :

[0028] ;

[0029] In the formula, It is inertial weight The maximum value, It is inertial weight The minimum value, The maximum number of iterations, This represents the current iteration number.

[0030] As a preferred embodiment of the present invention

[0031] The objective function defined in step 3 for:

[0032] ;

[0033] In the formula It is a specific angular frequency threshold.

[0034] The rotation matrix R in step S2 is:

[0035] .

[0036] As a preferred embodiment of the present invention

[0037] In step 2, the extreme grid impedance model corresponds to a short-circuit ratio (SCR) of 1.5, and the grid inductance... =0.0021H.

[0038] As a preferred embodiment of the present invention

[0039] The boundary constraint processing in step S4 is as follows:

[0040] If the position is outside the feasible region of the parameters, a boundary snapping strategy is used to set it as the corresponding boundary value.

[0041] As a preferred embodiment of the present invention

[0042] The dividing frequency for the full frequency band in step S2 is: =50Hz.

[0043] As a preferred embodiment of the present invention

[0044] The maximum number of iterations in step S5 is 50.

[0045] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] 1. This invention creatively introduces extended passivity theory, transforming the low-frequency admittance matrix through a rotation matrix. This converts the traditional passivity index, which is always negative and cannot be directly used for optimization, into a quantifiable, continuous, and physically meaningful positive index. This successfully avoids the inherent failure problem of traditional passivity theory in the low-frequency range, providing the optimization algorithm with a continuous, differentiable search space with clear physical meaning.

[0048] (2) This invention introduces an extreme weak grid impedance model as a boundary condition for stability margin assessment, enabling the optimization objective function to simultaneously consider high-frequency harmonic stability and low-frequency synchronization stability. This significantly improves the inverter's performance under normal operating conditions, giving the optimized inverter strong robustness to extremely weak grids. Furthermore, it maintains strong robustness and synchronization stability even when facing severe grid strength fluctuations and extremely weak grids, suppressing broadband oscillation risks from the design stage.

[0049] (3) This invention uses an improved adaptive particle swarm optimization algorithm to replace the traditional manual trial-and-error method, realizing efficient global optimization in a multi-dimensional, nonlinear, and strongly coupled control parameter space. By setting dynamic inertia weights, boundary adsorption mechanisms, and stability penalty functions, the algorithm can automatically and quickly search for the globally optimal or near-optimal parameter combination that maximizes the stability margin across the entire frequency band, significantly improving the efficiency, accuracy, and repeatability of parameter tuning. Attached Figure Description

[0050] Figure 1 This is a flowchart illustrating an inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention.

[0051] Figure 2 This is a flowchart illustrating the inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention.

[0052] Figure 3 This is a schematic diagram of the converter model control structure used in the inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention.

[0053] Figure 4 This is a schematic diagram of the phase-locked loop (PLL) control structure of the inverter model used in the inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention.

[0054] Figure 5 This is a schematic diagram of the power outer loop and current inner loop of the converter model control structure used in the inverter parameter optimization method based on passive theory and particle swarm algorithm proposed in this invention.

[0055] Figure 6The results show the verification of the mathematical and simulation models used in the inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention.

[0056] Figure 7 The results of the particle swarm optimization algorithm used in the inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm proposed in this invention are as follows:

[0057] Figure 8 This is a comparison chart of the low-frequency passive index of the inverter before and after optimization using the inverter parameter optimization method based on passiveity theory and particle swarm optimization proposed in this invention.

[0058] Figure 9 This is a comparison chart of the passive index of the inverter across the entire frequency band before and after optimization using the inverter parameter optimization method based on passive theory and particle swarm optimization proposed in this invention.

[0059] Figure 10 This paper presents simulation results of the inverter under weak grid conditions before and after optimization using an inverter parameter optimization method based on passivity theory and particle swarm optimization proposed in this invention. Detailed Implementation

[0060] The specific embodiments of the present invention are described below with reference to the accompanying drawings and examples:

[0061] 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. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0062] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0063] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0064] It should be noted that the structures, colors, proportions, sizes, etc. shown in the accompanying drawings are only used to complement the content disclosed in the specification, so that those skilled in the art can understand and read them, and are not intended to limit the conditions under which the present invention can be implemented. Any modifications to the structure, changes in the proportions, or adjustments to the size, without affecting the effects and objectives that the present invention can produce, should still fall within the scope of the technical content disclosed in the present invention.

[0065] like Figures 1-2 As shown, an inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm is particularly applicable to energy storage inverters, wind power converters, and photovoltaic converters, and includes the following steps:

[0066] Step S1: Construct a full-band system model of the inverter.

[0067] Obtain the circuit topology parameters and initial control parameters of the inverter, and establish a second-order small-signal admittance matrix model Y(s) of the inverter in the dq coordinate system. The admittance matrix model Y(s) includes the phase-locked loop, the inner current loop, the outer voltage loop, and the dynamics of the filter.

[0068] like Figure 3 As shown, this paper takes an energy storage converter as an example to conduct an analysis and constructs an admittance model for the energy storage converter. In order to focus on the interaction stability between the inverter's own control characteristics and the power grid, without analyzing the impact of the DC / DC link on the system stability, it is assumed that the DC equipment has a strong throttling capability and the dynamic influence of the DC side is ignored. In this case, the converter stability is dominated by the inverter characteristics. This assumption also applies to wind power and photovoltaic converters with similar control structures.

[0069] Energy storage and grid-based control are currently the mainstream control strategies. Their core function is to achieve efficient transmission of renewable energy power. Relying on grid voltage as a synchronization reference, they passively follow the system phase through a phase-locked loop (PLL) and can execute power commands. Their general control structure is as follows: Figure 4 and Figure 5 As shown, its working principle is as follows: The phase-locked loop (PLL) detects the grid connection point voltage in real time to obtain the grid synchronization phase; the power outer loop generates the reference value of the current inner loop according to the given active and reactive power commands; the current inner loop calculates the modulation voltage of the converter by quickly tracking the current reference value.

[0070] Based on small-signal linearization modeling, the inverter port admittance matrix is ​​derived:

[0071] Based on the small-signal linearization method, a complete system including a phase-locked loop, power outer loop, current inner loop, measurement delay, and LC filter is modeled, and its port admittance matrix is ​​derived. The mathematical models of each core control element are as follows:

[0072] (1) In the phase-locked loop synchronization stage, the grid-type energy storage adopts current vector decoupling control based on terminal voltage orientation, and the phase-locked loop detects the terminal voltage phase.

[0073] ;

[0074] In the formula, Represents the voltage at the grid connection point Axial components; Represents an intermediate variable; The phase angle of the phase-locked loop output; This represents the initial value of the angular velocity; Represents the angular velocity variable; and The corresponding control parameters for the phase-locked loop PI controller.

[0075] (2) Power outer loop: The power outer loop generates the reference value of the current inner loop by giving active and reactive power output through the upper-level command.

[0076] ;

[0077] and These are the active power and reactive power commands given by the superior authority, respectively. , These are the active power and reactive power obtained from actual measurements, respectively. , , and The control parameters of the outer loop PI controller corresponding to the active and reactive power of grid-connected energy storage; and Represents the intermediate variable of the differential. This represents the d-axis current reference value of the inner current loop under the grid-type control strategy. This represents the q-axis current reference value of the inner current loop under the grid-type control strategy.

[0078] (3) Current control inner loop: The current inner loop further generates the reference voltage of the converter through the reference value.

[0079] ;

[0080] and Represents an intermediate variable; and The current reference values ​​for the d-axis and q-axis are respectively. , , and The corresponding control parameters of the inner loop PI controller for grid-connected energy storage current; Represents the q-axis component of the current. Represents the filter inductor; Represents the d-axis component of the voltage; Represents the q-axis component of the voltage; and These are the d-axis and q-axis modulation reference voltages calculated and output by the controller, respectively.

[0081] Based on the established mathematical model, a step response was simulated in MATLAB and compared with the actual simulation waveform. The power reference value was suddenly increased by 0.01 pu, and the output change was observed. The results are as follows. Figure 6 As shown, the mathematical model is consistent with the electromagnetic transient simulation results.

[0082] The energy storage impedance model in the dq coordinate system is further generated based on the state-space model as follows:

[0083] ;

[0084] ;

[0085] In the formula, The matrix represents the state variables of the energy storage converter. Let A be a differential operator, and let A, B, and C be the coefficient matrices of the state space. The dq-axis components of the grid connection point voltage. For the dq-axis components of the grid-connected current, the matrix... Let be the admittance matrix in the dq coordinate system, whose elements are all functional expressions of the frequency s=jw and the control parameter x. This is a small-signal disturbance of the dq-axis component of the grid-connected current. For small-signal disturbances in the dq-axis components of the grid connection point voltage, It is an identity matrix.

[0086] The optimal control parameter x is selected as follows:

[0087] ;

[0088] Step S2: Define evaluation metrics using the extended passive theory.

[0089] The evaluation index is defined using extended passive theory. To address the failure of low-frequency indexes, a boundary frequency is set. The entire frequency band is divided into low-frequency band and high-frequency band.

[0090] In this embodiment, the dividing frequency for the entire frequency band is: =50Hz, where 50Hz is the standard power frequency (or fundamental frequency) of a power system. Using this as a dividing point allows us to physically separate the two different dynamics that dominate stability issues. The low-frequency band mainly reflects synchronization stability and subsynchronous oscillation problems related to the synchronization mechanism; the high-frequency band mainly covers harmonic and high-frequency resonance stability problems caused by controller bandwidth, filter resonance, etc. This division allows for the application of the most targeted analysis methods for the instability mechanisms of different frequency bands.

[0091] For the low-frequency band, a rotation matrix R from extended passive theory is introduced to perform a rotation transformation on the low-frequency admittance matrix. The eigenvalues ​​of the rotated admittance matrix are then calculated as stability indicators for the low-frequency band. .

[0092] In this embodiment, the rotation matrix R is selected as:

[0093] ;

[0094] Low-frequency stability index obtained for:

[0095] ;

[0096] In the formula, Represents different parameters and different frequencies The extended passivity index below; Represents the Hermitian transformation of the rotation matrix R; It is composed of different parameters and different frequencies The complex function formed by these components represents the converter admittance values ​​under different parameters and frequencies, and is a second-order matrix. Represents the converter admittance matrix The Hermitian transform.

[0097] Simultaneously calculate the low-frequency passive index of the grid impedance under the preset extreme grid impedance. :

[0098] ;

[0099] in G represents the grid impedance. Assuming the grid impedance is unknown, to ensure converter stability, the grid impedance is chosen as the extreme case, and G is set as:

[0100] ;

[0101] The low-frequency margin is defined as follows:

[0102] ;

[0103] For high-frequency bands, conventional passive performance indicators are used as stability indicators. The high-frequency stability margin is defined as follows:

[0104] .

[0105] In this embodiment, the extreme grid impedance model corresponds to a short-circuit ratio (SCR) of 1.5 and a grid inductance of [missing information]. =0.0021H. The grid inductance Lg=0.0021H is a specific physical parameter corresponding to SCR=1.5. SCR=1.5 represents a typical extremely weak grid condition, indicating an extreme operating condition where the grid strength at the grid connection point is severely insufficient. Using this as the boundary condition for optimization design means that the control parameters tuned by the method of this invention can not only meet the stable operation requirements under conventional grid conditions, but also remain stable when the grid strength drops drastically to a critical level. This ensures that the final parameters have grid adaptability, improving the inverter's adaptability and robustness to grid strength fluctuations.

[0106] Step S3: Construct the optimization objective function.

[0107] A full-band stability margin function is constructed as the optimization objective, and an improved adaptive particle swarm optimization algorithm is used to find the optimal function in the multi-dimensional control parameter space. First, the improved particle swarm optimization algorithm is used to define the objective function. For low-frequency margin With high-frequency margin The minimum value in the objective function The global minimum of the stability margin across the entire frequency band:

[0108] ;

[0109] In the formula, It is a specific angular frequency threshold.

[0110] This function exhibits constant negative stability information in the low-frequency range of conventional passive indices. Combined with the rotation matrix R in step S2, the objective function... This is transformed into quantifiable positive values, thus providing an effective gradient direction for the optimization algorithm. The optimization objective is to find the control parameter vector. ,make Maximize and exceed 0 within the given parameter range.

[0111] Step S4: Execute the particle swarm optimization algorithm.

[0112] Initialize the particle swarm, setting the population size, maximum number of iterations, learning factor, and inertia weights for the particle swarm optimization algorithm. Randomly generate an initial particle population containing multiple sets of control parameters. In each iteration, calculate the objective function value for each particle under the corresponding control parameters as its fitness, update the individual's historical best position (Pbest) and the global historical best position (Gbest), and use a dynamic inertia weight strategy to update the particle's velocity and position. Apply boundary constraints to the updated parameters to ensure they remain within the physically feasible region. Specifically, this includes:

[0113] Step S4.1: Set algorithm parameters, including population size, maximum number of iterations, learning factor, and inertia weight range. Randomly generate the initial position matrix X and initial velocity matrix V within the feasible region of the parameters. In this embodiment, the population size N=30, and the maximum number of iterations is set to... =50, feasible region is The participation factors are c1=c2=1.5. The maximum value of the inertia weight is set. =0.9 and minimum value =0.4 and adopt a linear decreasing strategy to dynamically adjust the inertia weight, so as to balance the global search capability of the algorithm in the early stage of iteration and the local development capability in the later stage of iteration.

[0114] Step S4.2: Dynamically adjust the inertia weight using a linear decreasing strategy. :

[0115] ;

[0116] In the formula, It is inertial weight The maximum value, It is inertial weight The minimum value, The maximum number of iterations, This represents the current iteration number.

[0117] During the iteration process, the fitness function value is calculated based on the current position of each particle; if the calculation result indicates that the system is unstable, a penalty function is introduced to assign the particle an extremely low fitness value. The maximum speed is limited based on the update speed of individual cognition and social cognition. Within the specified range, prevent particles from exceeding the effective solution space:

[0118] ;

[0119] ;

[0120] Step S4.3: Update the particle's velocity and position based on the individual historical best position (Pbest) and the global historical best position (Gbest), and apply amplitude clamping to the updated velocity and boundary constraints to the position:

[0121] ;

[0122] If the updated position exceeds the set boundary If the boundary adsorption strategy is adopted, it will be forcibly set to the corresponding boundary value.

[0123] In this embodiment, the handling of converter-related parameters and their optimization results are shown in Table 1: (Note: There are multiple solutions for the parameters that maintain positive resistance across the entire frequency band. The optimization results of the example used in this invention within a given range are as follows)

[0124] Table 1 Comparison of Optimized Control Parameters for Grid-connected Energy Storage Systems

[0125] parameter initial value Optimization value Parameter range Kppll / Kipll 50pu / 1000pu / s 20pu / 995pu / s [20,100] / [100,3000] Kp1 / Ki1 4pu / 10pu / s 1.2 PU / 55.28 PU / s [0.5,5] / [1,100] Kp2 / Ki2 4pu / 10pu / s 0.5 pu / 3.07 pu / s [0.5,5] / [1,100] Kp3 / Ki3 0.5 pu / 100 pu / s 2pu / 20pu / s [0.1,2] / [1,200] Kp4 / Ki4 0.5 pu / 100 pu / s 2pu / 20pu / s [0.1,2] / [1,200]

[0126] Step S5: Output the optimal parameters.

[0127] Repeat step S4 to iterate and approach the globally optimal solution that maximizes the stability margin across the entire frequency band. Determine if the maximum number of iterations has been reached. The algorithm checks whether the change in the global historical best position (Gbest) over several consecutive iterations is less than a threshold. If so, the iteration terminates, and the control parameter combination corresponding to the global historical best position (Gbest) is used as the final tuning parameters for the inverter. The particle swarm optimization algorithm is as follows: Figure 7 As shown.

[0128] In this embodiment, the maximum number of iterations in step S5 is 50. This ensures that the particle swarm optimization algorithm has sufficient iteration space for effective global exploration and local development, thereby increasing the probability of finding a stable and high-quality solution, while avoiding unnecessary consumption of computational resources and time costs due to excessive iterations.

[0129] After 50 iterations, the algorithm converges. It outputs the optimal parameter x. The passivity index is compared to... Figure 8 and Figure 9 As shown, after using this parameter, the inverter's rotating passivity index in the low-frequency range significantly improved from a negative value to a positive value, while remaining stable in the high-frequency range, verifying the effectiveness of this method. Further verification of the calculation results was performed in PSCAD / EMTDC. Figure 10 As shown.

[0130] A computer device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. The memory and the processor are communicatively connected. The memory stores computer instructions, and the processor executes the computer instructions to perform an inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm.

[0131] A computer program, such as computer program instructions, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the present invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0132] The preferred embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited to the above embodiments. Within the scope of knowledge possessed by those skilled in the art, various changes can be made without departing from the spirit of the present invention.

[0133] Many other changes and modifications can be made without departing from the concept and scope of this invention. It should be understood that this invention is not limited to the specific embodiments, and the scope of this invention is defined by the appended claims.

Claims

1. An inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm. Its features are, Includes the following steps: Step S1: Construct a full-band system model of the inverter. Obtain the circuit topology parameters and initial control parameters of the inverter, and establish a second-order small-signal admittance matrix model Y(s) of the inverter in the dq coordinate system. The admittance matrix model Y(s) includes a phase-locked loop, an inner current loop, an outer voltage loop, and filter dynamics. Step S2: Define evaluation metrics using the extended passive theory. Set the boundary frequency The entire frequency band is divided into low-frequency and high-frequency bands. For the low-frequency band, a rotation matrix R from extended passive theory is introduced to perform a rotation transformation on the low-frequency admittance matrix. The eigenvalues ​​of the rotated admittance matrix are then calculated as stability indicators for the low-frequency band. For high-frequency bands, conventional passive performance indicators are used as stability indicators for high-frequency bands. Based on the pre-defined extreme grid impedance model, low-frequency margins are constructed respectively. and high-frequency margin ; Step S3: Construct the optimization objective function. Define the objective function For low-frequency margin With high-frequency margin The minimum value in the objective function To find the global minimum of the stability margin across the entire frequency band, and combining it with the rotation matrix R in step S2, the objective function is... The result is transformed into quantifiable positive values, and the optimization objective is to find the control parameter vector. ,make Maximize and exceed 0 within the given parameter range; Step S4: Execute the particle swarm optimization algorithm. Set the population size, maximum number of iterations, learning factor, and inertia weight of the particle swarm optimization algorithm. Randomly generate an initial particle population containing multiple sets of control parameters. In each iteration, calculate the objective function value of each particle under the corresponding control parameters as the fitness. Update the individual historical best position and the global historical best position. Use a dynamic inertia weight strategy to update the velocity and position of the particles and perform boundary constraint processing on the updated parameters. Step S5: Output the optimal parameters. Repeat step S4 until the fitness value meets the convergence condition, and output the combination of control parameters corresponding to the global optimal position as the final tuning parameters of the inverter.

2. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, The low-frequency stability index in step S2 The calculation formula is: ； In the formula, Represents different parameters and different frequencies The extended passivity index below; Represents the Hermitian transformation of the rotation matrix R; It is composed of different parameters and different frequencies The complex function formed by these components represents the converter admittance values ​​under different parameters and frequencies, and is a second-order matrix. Represents the converter admittance matrix The Hermitian transform.

3. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1. In step S4, a linear decreasing strategy is used to dynamically adjust the inertia weight. : ； In the formula, It is inertial weight The maximum value, It is inertial weight The minimum value, The maximum number of iterations, This represents the current iteration number.

4. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, The objective function defined in step 3 for: ； In the formula It is a specific angular frequency threshold.

5. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, The rotation matrix R in step S2 is: 。 6. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, In step 2, the extreme grid impedance model corresponds to a short-circuit ratio (SCR) of 1.5, and the grid inductance... =0.0021H.

7. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, The boundary constraint processing in step S4 is as follows: If the position is outside the feasible region of the parameters, a boundary snapping strategy is used to set it as the corresponding boundary value.

8. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1, Its features are, The dividing frequency for the full frequency band in step S2 is: =50Hz.

9. The inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in claim 1. Its features are, The maximum number of iterations in step S5 is 50.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the inverter parameter optimization method based on passivity theory and particle swarm optimization algorithm as described in any one of claims 1-9.

Images