Network-configuration-type converter grid impedance identification method and system based on load sudden transient characteristics

Abstract

The application discloses a network-constructing type converter power grid impedance identification method and system based on load mutation transient characteristics, obtains a grid-connected point voltage and a load branch current; calculates real active power of a local load, and obtains a real active power tracking curve; calculates an active power derivative according to the real active power; judges that a load mutation occurs according to the active power derivative, and obtains load active power at a moment before the load mutation and load active power at a moment after the load mutation; and obtains a load mutation step amount through first-order difference; obtains an overshoot amount of the jump peak value according to the real active power tracking curve; calculates a large power grid equivalent line inductance according to the load mutation step amount and the overshoot amount, and equivalent short-circuit impedance is determined by the large power grid equivalent line inductance. Through acquisition of active power transient characteristics caused by load switching, online non-intrusive identification of power grid equivalent impedance / inductance is carried out, and power grid impedance can be accurately identified without affecting power quality of the power grid.

Description

Technical Field

[0001] This invention relates to power electronics and power system control, specifically to a method and system for identifying grid impedance of a grid-type converter based on the transient characteristics of load changes. Background Technology

[0002] With the acceleration of the global energy transition, the penetration rate of distributed renewable energy sources such as wind and solar power in the power grid is constantly rising. Modern power systems are undergoing a profound evolution from being dominated by traditional synchronous generators to being dominated by power electronics. Because power electronic converters lack the physical rotational inertia of traditional generators, and the system's short-circuit capacity decreases significantly, the power grid exhibits increasingly pronounced characteristics of a "weak grid" or even an "extremely weak grid." In a weak grid environment, the proportion of the equivalent impedance of lines in the total system impedance rises sharply. In actual distribution network or microgrid operation, with daily network topology reconfiguration (such as line disconnection / connection), grid-connected transformer switching, and real-time fluctuations in tidal loads, the short-circuit ratio (SCR) at the point of connection (PCC) undergoes frequent, drastic, and unpredictable dynamic changes.

[0003] Grid-connected converters are considered a core technology for solving the stability of weak power grids because they can actively establish the voltage and frequency at the grid connection point. However, grid-connected converters are essentially controlled voltage sources, and their output active and reactive power are extremely sensitive to the external grid impedance. There is a strong nonlinear coupling between the core control parameters inside the converter (such as virtual inertia J, damping coefficient D, and voltage droop coefficient) and the actual impedance parameters of the large power grid. If the converter controller operates with fixed preset parameters, when the external grid impedance increases sharply, the original control parameters will no longer match, the system damping ratio will deteriorate rapidly, and broadband subsynchronous oscillations, low-frequency power swings, and even large-scale cascading grid disconnection accidents can easily occur between the converter and the weak grid. Therefore, obtaining the equivalent impedance of the power grid has become a problem that needs to be solved.

[0004] Existing methods for online detection of grid impedance are mainly divided into active disturbance injection and passive observation methods, but both have drawbacks. The active disturbance method calculates impedance by controlling the inverter to actively inject non-characteristic disturbance signals into the grid. Although the identification accuracy is acceptable, its "intrusive" nature directly increases the total harmonic distortion of the system, pollutes the grid environment, and makes it difficult to meet stringent power quality standards. Moreover, in complex weak grids containing multiple converters, artificially injected disturbances are prone to nonlinear coupling with the control loops of surrounding equipment, inducing unpredictable broadband resonances and posing serious safety hazards.

[0005] To avoid grid pollution caused by active injection, passive identification methods based on phase-locked loops (PLLs) or state observers have been proposed in recent years. These methods estimate impedance by capturing minute changes in the "micro-phase" of the grid voltage, but they are prone to failure in real, weak grids with low short-circuit ratios. On one hand, the PCC point voltage in weak grids is extremely vulnerable, accompanied by numerous background harmonics and distortions. Microscopic observations based on voltage phase differentiation naturally amplify high-frequency noise, easily leading to severe fluctuations or even divergence in the identification results. On the other hand, purely mathematical observers lack fault detection capabilities, making it difficult to distinguish between genuine network impedance changes and external short-circuit faults. When a remote fault occurs in the grid, they are prone to outputting incorrect SCR (Short-Circuit Reduction) estimates, potentially causing system malfunctions.

[0006] In summary, existing technologies consistently fail to achieve an effective balance between complete non-intrusion and high noise robustness, struggling to address both "zero disturbance contamination" and "high noise robustness." Active injection methods can disrupt power quality and induce resonance, while traditional passive observation methods rely excessively on easily distorted microscopic voltage phases, making them highly susceptible to high-frequency noise interference and failure in weak power grids. Online identification methods suffer from severe time lag, making it difficult to achieve rapid tracking during sudden changes in grid impedance. Traditional online identification algorithms (such as spectral analysis and recursive iterative methods) typically rely on long data time windows or low-pass filtering stages, resulting in significant output delays. When a large power grid experiences a sharp drop in the short-circuit ratio (SCR), the sluggish identification results cannot provide timely damping adaptive matching for grid-connected converters, easily inducing system transient instability. Summary of the Invention

[0007] Purpose of the invention: To address the above-mentioned shortcomings, this invention provides a non-intrusive, stable, and fast method and system for identifying grid impedance of grid-type converters based on the transient characteristics of load changes.

[0008] Technical Solution: To solve the above problems, the present invention provides a method for identifying grid impedance of a grid-type converter based on the transient characteristics of load changes, comprising the following steps:

[0009] Obtain the grid connection point voltage and load branch current to obtain the true active power tracking curve of the inverter side of the grid-connected converter; calculate the true active power of the local load.

[0010] Calculate the active derivative based on the actual active power of the local load;

[0011] When a load change occurs, the active power of the load is obtained at the moment before the load change and at the moment after the load change, based on the active power derivative. The load change step is obtained by performing a first-order difference between the active power of the load before and after the load change.

[0012] The overshoot of the jump peak is obtained from the actual active power tracking curve;

[0013] The equivalent line inductance of the large power grid is calculated based on the load abrupt change step and overshoot. The equivalent short-circuit impedance is determined by the equivalent line inductance of the large power grid. The calculation formula is:

[0014] ;

[0015] in, For filtering inductors, This represents the step change in load. This is the overshoot.

[0016] Furthermore, the load branch current is obtained directly through a sensor, or by obtaining the inverter-side output current and the grid-side interactive current, and calculating the difference between the inverter-side output current and the grid-side interactive current to obtain the load branch current.

[0017] Furthermore, in the rotating dq coordinate system, the actual active power of the local load is calculated using real-time measured grid connection point voltage and load branch current components through instantaneous power theory. The calculation formula is:

[0018] ;

[0019] in, The d-axis component of the grid connection point voltage. This represents the d-axis component of the load branch current. The q-axis component of the grid connection point voltage. This represents the q-axis component of the load branch current.

[0020] Furthermore, when the active derivative exceeds the threshold, a load mutation is determined to have occurred.

[0021] Furthermore, the short-circuit ratio is calculated based on the load abrupt change step and overshoot. The calculation formula is as follows:

[0022] ;

[0023] in, For short-circuit ratio, The rated line voltage at the grid connection point, For the rated capacity of the grid-type converter, This is the fundamental angular frequency of the power grid.

[0024] Furthermore, the formula for calculating the equivalent short-circuit impedance is as follows: .

[0025] The present invention discloses a grid impedance identification system for grid-type converters based on load transient characteristics, comprising:

[0026] The data acquisition module is used to acquire grid connection point voltage and load branch current to obtain the actual active power tracking curve on the inverter side of the grid-connected converter.

[0027] The active power calculation module is used to calculate the actual active power of the local load; calculate the active power derivative based on the actual active power of the local load; and obtain the overshoot of the peak value based on the actual active power tracking curve.

[0028] The event trigger detection module is used to determine the occurrence of load change based on the active derivative; and to obtain the load active power at the moment before the load change and the load active power at the moment after the load change; the load active power at the moment before the load change and the moment after the load change are differentially expressed to obtain the load change step.

[0029] Grid impedance identification is used to calculate the equivalent line inductance of a large power grid based on load abrupt changes and overshoot. The equivalent short-circuit impedance is determined by the equivalent line inductance of the large power grid. The calculation formula is:

[0030]

[0031] in, For filtering inductors, This represents the step change in load. This is the overshoot.

[0032] Furthermore, a grid-type converter grid includes grid-type converters, local filter networks, grid connection points, local loads, and the equivalent impedance of the main grid.

[0033] The present invention provides an electronic device comprising a processor and a memory; the processor executes the steps of the above method by calling a program or instruction stored in the memory.

[0034] The present invention provides a computer-readable storage medium for storing a program or instructions that cause a computer to perform the steps of the above-described method.

[0035] In actual microgrid or distribution network operation, the daily switching of local loads (such as the starting of large motors, sudden addition or removal of large-capacity loads) is an extremely common and unavoidable natural disturbance. When the load changes abruptly, a significant transient power redistribution occurs between the grid-connected converter and the grid based on the electrical distance. However, existing technologies merely treat this load change as a "disturbance" that needs to be passively mitigated, completely ignoring the enormous informational value hidden behind this natural transient process. In fact, the macroscopic time-domain trajectory of the converter's active power waveform after being disturbed (such as instantaneous peak jumps and damped oscillation periods) is strictly determined by the distribution of the system's closed-loop poles, and these poles are related to the unknown impedance of the external large power grid. There exists a direct and unique mapping relationship. More importantly, due to the inherent virtual inertia of the grid-type converter system... It has natural low-pass filtering characteristics, and the macroscopic dynamic waveform of active power can spontaneously filter out high-frequency voltage noise.

[0036] Beneficial effects: Compared with the prior art, the significant advantage of this invention is that by acquiring the transient characteristics of active power caused by load switching, it can perform online non-intrusive identification of the grid equivalent impedance / inductance, which can accurately identify the grid impedance without affecting the power quality of the grid. Attached Figure Description

[0037] Figure 1 This is a block diagram of the control system for the grid-type converter in this invention.

[0038] Figure 2 This is the active power tracking curve under sudden load in this invention.

[0039] Figure 3 This is a schematic diagram of the surface distribution of overshoot, load mutation, and short-circuit ratio in three-dimensional space in this invention. Detailed Implementation

[0040] like Figure 1 As shown, in a specific embodiment of the present invention, the system to be identified consists of a grid-type converter (using VSG control) and a local filter network ( ), local filter capacitor ( ), PCC point, local load and equivalent impedance of the main power grid ( It consists of ) . The virtual inertia of the VSG controller (Virtual Synchronous Generator Controller) is . Virtual damping is The fundamental angular frequency is The three-phase voltage on the grid side is , , Through the cascaded coordination of the outer and inner loops, precise regulation of the active power, AC voltage, and output current of the grid-connected system is achieved. In specific operation, the system inverter side outputs three-phase voltage ( , , ) and three-phase current ( , , The underlying physical quantities are first obtained by sampling the three-phase current on the grid side through sensors. , , The three-phase AC signals are then mapped to a rotating coordinate system via an abc / dq0 coordinate transformation to obtain the active power feedback. and voltage amplitude feedback quantity These feedback signals enter the outer loop of the VSC controller and are respectively compared with the system's set active power reference value. AC voltage reference value Deviation calculations are performed, and after processing by the frequency regulator and voltage regulator, the angle command value required for inner-loop current control is generated. With dq axis voltage command value Subsequently, the voltage and current dual closed-loop control module, which incorporates grid phase information, inputs this information in conjunction with the actual d-axis and q-axis currents acquired and transformed in real time from the grid side. and capacitor terminal voltage and the dq current at the inverter output. High-bandwidth closed-loop comparison and decoupling operations are performed to ultimately output a three-phase voltage modulation wave command. The instruction is sent to the SPWM modulation stage and directly converted into a pulse signal to control the switching of the main circuit transistors. This drives the VSC bridge arm to achieve power conversion between the DC and AC sides, thereby ensuring that the actual output voltage and current of the system closely follow the given instruction, achieving the purpose of stable grid connection or independent power supply.

[0041] This embodiment presents a grid impedance identification method for grid-type converters based on load transient characteristics, comprising the following steps:

[0042] Obtain the grid connection point voltage and load branch current to obtain the true active power tracking curve of the inverter side of the grid-connected converter; calculate the true active power of the local load.

[0043] Calculate the active derivative based on the actual active power of the local load;

[0044] Based on the active derivative, when a load change occurs, obtain the load active power at the moment before the load change and the load active power at the moment after the load change; perform a first-order difference between the load active power at the moment before and the moment after the load change to obtain the load change step.

[0045] The overshoot of the jump peak is obtained from the actual active power tracking curve;

[0046] The equivalent line inductance of the large power grid is calculated based on the load change step and overshoot. The equivalent short-circuit impedance is derived from the equivalent line inductance of the large power grid.

[0047] For a power grid system, by performing small-signal linearization and Laplace transform at the steady-state operating point, the power angle response equation of the VSG controller can be obtained:

[0048]

[0049] In the formula, For the infinitesimal increase in electromagnetic power, For the incremental change of the work angle, is the complex frequency variable in the Laplace transform.

[0050] In a grid-connected inverter, it is assumed that the VSG is equivalent to a circuit with an inductive output impedance (i.e., The voltage source is connected in parallel to the grid voltage. The location. Its output active power. It satisfies the classical transport equation:

[0051]

[0052] In the formula, This represents the total system reactance. VSG internal potential With grid voltage The angle of attack between them.

[0053] Active power Optical angle The partial derivatives of the synchronous power coefficient are In small-signal analysis, it represents the system at its equilibrium point. Nearby, the amount of power recovery produced by a unit change in power angle:

[0054]

[0055] Substituting equation (1.2) into the equation, we get:

[0056]

[0057] At steady-state operating point At that point, we obtained:

[0058]

[0059] Synchronous power factor Directly reflects the equivalent impedance of the unknown power grid The two factors have a strongly coupled inverse relationship in terms of their impact on system stiffness.

[0060] When local load occurs at PCC point During a sudden change in active power, at the initial transient instant (before the rotor angle is constrained by inertia and a mechanical response occurs), the VSG is forced to bear the transient power increment. Determined by the electrical distance (impedance voltage division) within the system:

[0061]

[0062] In summary, the actual incremental change in total output active power detected by the VSG includes both steady-state regulation components and transient shunt components caused by power angle offset:

[0063]

[0064] The internal state variables in the above formula Eliminating these factors allows us to obtain the closed-loop transfer function of the output active power of the grid-type converter to the local load disturbance. :

[0065]

[0066] For step disturbance By performing the inverse Laplace transform, the system in the underdamped state can be obtained ( The time-domain analytical response expression for active power is as follows:

[0067]

[0068] In the formula, the system damping ratio Damped oscillation angular frequency Phase angle .

[0069] Based on the above time-domain analytical solution, this invention proposes the following feature extraction and impedance inverse estimation criteria to achieve the short-circuit ratio (SCR) or grid inductance (…). Online non-intrusive identification:

[0070] According to the initial value theorem, at the instant the local load is added or removed ( The first transient peak value of active power. satisfy:

[0071]

[0072] Once this overshoot is obtained, its characterization relationship with the Short-Circuit Ratio (SCR) can be established. The definition of the Short-Circuit Ratio (SCR) is...

[0073]

[0074] in This indicates the rated apparent power (VA) of the grid-type converter. It represents the short-circuit capacity (VA) that the system can provide when a three-phase symmetrical short circuit occurs at the PCC point (grid connection point).

[0075] According to power system theory, the short-circuit capacity of the power grid Determined by the grid voltage and equivalent short-circuit impedance:

[0076]

[0077] in: This is the effective value of the rated line voltage at point PCC. This is the equivalent short-circuit impedance of the large power grid.

[0078] Because high-voltage power grids or microgrid lines are typically significantly inductive. Ignoring line resistance, the equivalent short-circuit impedance is entirely determined by the equivalent line inductance of the large power grid to be identified. Decide:

[0079]

[0080] in The fundamental angular frequency of the power grid (e.g., 50Hz corresponds to) (rad / s).

[0081] Substituting formula (1.13) into formula (1.12), we obtain the short-circuit capacity and... Direct relationship:

[0082]

[0083] From equation (1.10) we can derive formula:

[0084]

[0085] Substituting these values ​​into the short-circuit capacity formula (1.14), and then into the SCR definition formula (1.11), we can obtain the final analytical expression for directly calculating the short-circuit ratio from the active power jump peak:

[0086]

[0087] Let's make a slight adjustment to formula (1.16) to make its physical meaning clearer:

[0088]

[0089] In the formula, The rated line voltage at the grid connection point. This represents the rated capacity of the converter. The mapping equation shows that the controller does not need to perform complex complex calculus calculations; it only needs to extract the initial peak value of the load change. By combining the known hardware constant coefficients, the current true SCR assessment value of the power grid can be output.

[0090] Figure 3 This describes the surface distribution characteristics of the mapping formula (1.17) in three-dimensional space. (From...) Figure 3 It can be seen that, in three-dimensional space, the same active power overshoot... (i.e., the same horizontal section on the Z-axis) corresponds to a line with different SCRs and load abrupt changes. The contour lines formed by combining elements. This means that if only the active power overshoot of the inverter itself is observed... If we reverse-engineer the SCR, the system will face spatial ambiguity and may obtain misclassified solutions that are very different from each other.

[0091] Therefore, in order for equation (1.17) to output a unique and accurate SCR setpoint, the true mutation amount of the local natural load must be obtained synchronously and independently within the controller. Therefore, based on the inverter's locally sampled electrical quantities and Kirchhoff's power conservation, It can be calculated, and the specific steps are as follows:

[0092] In the grid-connected point-of-charge (PCC) control architecture of modern grid-connected converters, the inverter's digital signal processor (DSP) typically performs synchronous high-frequency sampling of the inverter-side output current. Grid connection point voltage and grid-side interactive current (Or directly sample the load branch current). According to Kirchhoff's Current Law (KCL), at the PCC node, the instantaneous current vector flowing into the local load branch is... Strictly meet:

[0093]

[0094] In the rotating dq coordinate system, the inverter's main control chip can use its real-time measured voltage and current components to rapidly calculate the actual active power of the local load through instantaneous power theory. :

[0095]

[0096] in, The d-axis component of the grid connection point voltage. This represents the d-axis component of the load branch current. The q-axis component of the grid connection point voltage. This represents the q-axis component of the load branch current.

[0097] When the event triggering module detects the active derivative When a load change is detected due to an out-of-limit condition, the controller immediately locks the steady-state time prior to the change in memory. Total load power Simultaneously, at the microsecond instant that the load completes electrical connection ( Extract new load power By performing a first-order difference calculation on both, an extremely accurate load abrupt change step can be obtained:

[0098]

[0099] At this point, It was calculated. The SCR can be precisely estimated using Equation (1.17) by reading the active power tracking curve on the control side of the grid-connected inverter.

[0100] Experimental verification was conducted on the solution proposed in this invention. Firstly, an experimental setup was constructed... Figure 1Using an identical simulation model, the SCR (Short-Circuit Response Rate) is set by inputting the load, the rated power of the model, and the three-phase short-circuit specific capacity. The overshoot is obtained through simulation, and the formula (1.17) is used to verify the overshoot based on the set SCR and the obtained overshoot. For example, if the rated power of the model is set to 500kW and the three-phase short-circuit specific capacity is set to 1000KVA, then SCR = 20. Overshoot and SCR calculations are performed according to the set step size and value range. The SCR value ranges from 2 to 10 with a step size of 1; the load value ranges from 0 to 500kW with a step size of 50kW. The corresponding active power curve on the control side of the grid-connected inverter is shown below. Figure 2 As shown, the overshoot obtained from the simulation is the same as the overshoot calculated based on the known SCR using equation (1.17), which is sufficient to prove the effectiveness and accuracy of the scheme.

[0101] This invention eliminates the need for the inverter to actively inject any high-frequency harmonics, interharmonics, or negative-sequence currents into the power grid. It relies entirely on naturally occurring local load changes within the grid to trigger the transient response, fundamentally eliminating the risk of broadband resonance caused by artificially injected disturbances and meeting the most stringent grid-connected power quality standards. Furthermore, the analytical identification scheme of this invention directly reads the overshoot from the active power tracking curve on the grid-connected inverter control side to obtain the first-wave jump peak value. It eliminates the need to wait for the waveform to reach steady state or undergo lengthy integration observations. Within one transient fluctuation cycle, it can rapidly output the true SCR using extremely simple algebraic multiplication and addition operations, consuming minimal computing power from the DSP control chip.

Claims

1. A method for identifying grid impedance of a grid-type converter based on the transient characteristics of load mutation, characterized in that, Includes the following steps: Obtain the grid connection point voltage and load branch current to obtain the true active power tracking curve of the inverter side of the grid-connected converter; calculate the true active power of the local load. Calculate the active derivative based on the actual active power of the local load; When a load change occurs, the active power of the load is obtained at the moment before the load change and at the moment after the load change, based on the active power derivative. The load change step is obtained by performing a first-order difference between the active power of the load before and after the load change. The overshoot of the jump peak is obtained from the actual active power tracking curve; The equivalent line inductance of the large power grid is calculated based on the load abrupt change step and overshoot. The equivalent short-circuit impedance is determined by the equivalent line inductance of the large power grid. The calculation formula is: in, For filtering inductors, This represents the step change in load. This is the overshoot.

2. The grid impedance identification method for grid-type converters based on load transient characteristics according to claim 1, characterized in that, The load branch current is obtained directly by a sensor, or by obtaining the inverter-side output current and the grid-side interactive current, and then calculating the difference between the inverter-side output current and the grid-side interactive current to obtain the load branch current.

3. The grid impedance identification method for grid-type converters based on load transient characteristics according to claim 2, characterized in that, In the rotating dq coordinate system, the actual active power of the local load is calculated using real-time measured grid connection point voltage and load branch current components through instantaneous power theory. The calculation formula is: ； in, The d-axis component of the grid connection point voltage. This represents the d-axis component of the load branch current. The q-axis component of the grid connection point voltage. This represents the q-axis component of the load branch current.

4. The grid impedance identification method for grid-type converters based on load transient characteristics according to claim 3, characterized in that, When the active derivative exceeds the threshold, a load mutation is determined to have occurred.

5. The grid impedance identification method for grid-type converters based on load transient characteristics according to claim 4, characterized in that, The short-circuit ratio is calculated based on the load abrupt change step and overshoot. The formula is as follows: ； in, For short-circuit ratio, The rated line voltage at the grid connection point, For the rated capacity of the grid-type converter, This is the fundamental angular frequency of the power grid.

6. The grid impedance identification method for grid-type converters based on load transient characteristics according to claim 5, characterized in that, The formula for calculating the equivalent short-circuit impedance is as follows: .

7. A grid impedance identification system for a grid-type converter based on the transient characteristics of load abrupt changes, characterized in that, include: The data acquisition module is used to acquire grid connection point voltage and load branch current to obtain the actual active power tracking curve on the inverter side of the grid-connected converter. The active power calculation module is used to calculate the actual active power of the local load; calculate the active power derivative based on the actual active power of the local load; and obtain the overshoot of the peak value based on the actual active power tracking curve. The event trigger detection module is used to determine the occurrence of load change based on the active derivative; and to obtain the load active power at the moment before the load change and the load active power at the moment after the load change; the load active power at the moment before the load change and the moment after the load change are differentially expressed to obtain the load change step. Grid impedance identification is used to calculate the equivalent line inductance of a large power grid based on load abrupt changes and overshoot. The equivalent short-circuit impedance is determined by the equivalent line inductance of the large power grid. The calculation formula is: in, For filtering inductors, This represents the step change in load. This is the overshoot.

8. The grid impedance identification system for grid-type converters based on load transient characteristics according to claim 7, characterized in that, A grid-type converter power grid includes grid-type converters, local filter networks, grid connection, local loads, and the equivalent impedance of the main power grid.

9. An electronic device, characterized in that, Including processor and memory; The processor executes the steps of the method as described in any one of claims 1 to 6 by invoking programs or instructions stored in the memory.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a program or instructions that cause a computer to perform the steps of the method as described in any one of claims 1 to 6.

Images