Response-driven new energy AC / DC power grid voltage instability identification method and system

Abstract

The present invention proposes a response-driven voltage instability identification method and system for new energy AC / DC power grids. By analyzing the mechanisms and manifestations of different voltage instability scenarios, based on the spectrum clustering of reactive power flow, the partition section is determined based on the principle of the most drastic voltage change, and the measurable key response bus is determined. A correction function is introduced, and the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the key response bus voltage time series signal is selected as the characteristic quantity. The voltage instability scenario is identified according to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation. The new voltage instability scenario caused by the control system in the new energy AC / DC power grid can be distinguished from the traditional load-side voltage instability scenario, and the cause of the voltage instability can be accurately identified. The identification method is based on the power grid response information and does not need to rely on the system model, which is conducive to the implementation of non-plan strategies.

Description

Technical Field

[0001] The present invention belongs to the technical field related to power grids, and in particular to a response-driven new energy AC / DC power grid voltage instability identification method and system. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] With the large-scale grid connection of new energy and the commissioning of a large number of high-voltage direct current transmission projects, the transient voltage stability problem has become increasingly complex, resulting in a variety of new transient voltage instability scenarios. There have been many studies on the voltage instability scenarios on the traditional AC load side, and relatively complete identification methods and response strategies have been established. This relatively simple and single voltage instability problem can be dealt with through a plan-based strategy. However, for AC / DC hybrid power grids with a high proportion of new energy access, uncertainty makes it impossible to exhaustively enumerate the system operation modes, which undermines the premise of the establishment of the plan-based control strategy. In addition, the large number of power electronic devices included in the new energy access and high-voltage direct current transmission system have transient characteristics that are highly dependent on the control system, which also changes the traditional transient process dominated by the motor rotor equation, introduces new instability scenarios for the voltage stability problem, and further increases the risk of failure of the plan-based method. Therefore, it is necessary to implement safe and stable control based on real-time response information. Accordingly, the identification of voltage instability scenarios should also be transformed into a response-driven method.

[0004] On the other hand, the power system voltage stability assessment often uses the time domain simulation method or the direct method. Among them, the time domain simulation method requires the numerical solution of the high-dimensional differential algebraic equations of the power system to obtain the voltage curve of each bus over time, and further evaluate the transient voltage stability characteristics based on the curve. However, this method requires mathematical models of each device in the system, and requires the initial state of the system operation and disturbance information. The calculation amount is large and the accuracy depends on the accuracy of the model and parameters. It is mainly suitable for offline simulation and is difficult to adapt to the needs of rapid identification and assessment of voltage stability in complex power systems. The direct method quickly judges the transient stability of the power system through kinetic energy and potential energy in the transient process. However, the direct method has been proven to be able to only construct a more conservative energy function.

[0005] The Lyapunov index based on chaos theory only requires time domain information of electrical quantities, which is more in line with the needs of transient voltage stability identification based on response, and has become one of the feasible solutions for rapid voltage stability assessment. In previous studies on judging power system voltage stability based on Lyapunov index, some scholars considered the failure of DC continuous commutation failure, established an AC / DC hybrid system switch model, and evaluated transient voltage stability, but this method still relies on the system model to calculate the Lyapunov index, and for complex system models, the difficulty of solving it is still relatively large. Some other scholars have proposed a Lyapunov index calculation method using a finite voltage time series, which only requires the collection of voltage data for a limited time through a synchronous phasor measurement device for calculation, and can realize response-based transient voltage stability assessment without the need to establish a detailed system model. However, this method has defects: it requires dynamic response data of the entire network, which is difficult to achieve for large nonlinear systems; it can only judge voltage instability, but cannot identify different scenarios and mechanisms of voltage instability, and there is a phenomenon of missed judgment for pure voltage instability scenarios.

[0006] From the above analysis, it can be seen that the response-driven identification of voltage instability scenarios of new energy AC / DC power grids is essentially based on the traditional AC load-side voltage instability, taking into account the new instability scenarios brought by DC and new energy control systems. Under the traditional load-side voltage instability assessment system, considering new voltage instability scenarios will require upgrading the traditional voltage stability assessment system. On the other hand, it is necessary to propose targeted identification methods based on the characteristic differences between the load-side voltage instability scenario and the control system-dominated voltage instability scenario. Summary of the invention

[0007] In order to overcome the shortcomings of the above-mentioned prior art, the present invention provides a response-driven new energy AC / DC power grid voltage instability identification method and system, by analyzing the mechanisms of different voltage instability scenarios, selecting the anti-Lyapunov exponent zero-crossing fluctuation improvement index of the bus voltage timing signal as the key response characteristic quantity, and identifying the voltage instability scenarios according to the change law of the anti-Lyapunov exponent zero-crossing fluctuation improvement index, distinguishing the new voltage instability scenarios caused by the control system and the voltage instability scenarios caused by the load in the new energy power system.

[0008] To achieve the above-mentioned object, a first aspect of the present invention provides a response-driven new energy AC / DC power grid voltage instability identification method, comprising:

[0009] Implement voltage partitioning based on reactive power flow, determine the partition cross section based on the principle of the most drastic voltage change, and thus determine the key response bus that can be measured;

[0010] The voltage timing signal on the key response bus is obtained, a correction function is introduced to prevent the Lyapunov exponent from fluctuating at the zero crossing point due to an error, and an improvement index for preventing the Lyapunov exponent from fluctuating at the zero crossing point of the voltage timing signal on the key response bus is calculated;

[0011] According to the variation law of the improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from crossing zero point fluctuation, the scene identification result of the voltage instability caused by the control system or the voltage instability on the load side is obtained.

[0012] Furthermore, voltage partitioning is implemented based on reactive power flow, and the partition section is determined based on the principle that the bus voltage at both ends of the line changes most dramatically, so as to determine the key response bus that can be measured, specifically:

[0013] The bus voltage correlation matrix is ​​constructed based on the reactive power flow transmitted on the line to characterize the degree of bus voltage change at both ends of the line;

[0014] The bus voltage correlation matrix is ​​clustered by spectral clustering algorithm to obtain k1 partitions, the cross-section of each partition and the bus number contained in each partition, where k1 represents the expected number of partitions;

[0015] The lines on both sides of the partition section are key response buses that can be measured.

[0016] Furthermore, the bus voltage correlation matrix is ​​determined according to the inverse of the reactive power flow between the two buses, which is:

[0017] W=[w ij ] n×n ,

[0018] Where n is the number of buses, w ij It is the inverse of the reactive power flow between busbars i and j.

[0019] Furthermore, the voltage timing signal on the key response bus is taken, and a correction function is introduced to prevent the Lyapunov exponent from fluctuating at zero crossing due to errors. The improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from fluctuating at zero crossing is calculated, which is specifically:

[0020] Calculate the voltage amplitude deviation signal on each key response bus according to the acquired voltage timing signal on each key response bus and the determined voltage reference value;

[0021] Based on the voltage amplitude deviation signal, a correction function for preventing the Lyapunov exponent from fluctuating at zero crossing due to an error is obtained, and the correction function is used to calculate an improvement index for preventing the Lyapunov exponent from fluctuating at zero crossing at each moment.

[0022] The correction function is:

[0023]

[0024] In the formula, is the correction function, ξ is a very small positive value, A is the error coefficient, Δv (m-1)Δt is the voltage amplitude deviation of the key bus at time (m-1)Δt, v mΔt is the voltage amplitude on the key response bus at time mΔt; v' max is the maximum voltage amplitude change rate from fault removal to mΔt, where Δt is the number of sampling intervals of voltage time series data.

[0025] Furthermore, the correction function is used to calculate the improvement index for preventing the zero-crossing fluctuation of the Lyapunov exponent at each moment, specifically:

[0026]

[0027] Where λ1(kΔt) is the improvement index for preventing Lyapunov exponent zero-crossing fluctuation at time kΔt on the critical response bus; NΔt is the time window length, v B is the voltage reference value, v mΔt is the voltage amplitude of the key response bus at time mΔt, is the correction function.

[0028] Furthermore, according to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the key response bus voltage timing signal, the scenario identification results of voltage instability caused by the control system or load-side voltage instability are obtained, specifically: if the calculated anti-Lyapunov exponent improvement index is greater than zero and the duration of being greater than zero is greater than the set time threshold, and the key response bus voltage is less than the voltage threshold value, it is judged as a load-side voltage instability scenario; if the calculated anti-Lyapunov exponent improvement index repeatedly crosses the zero point, it is judged as a voltage instability scenario caused by the control system.

[0029] A second aspect of the present invention provides a response-driven new energy AC / DC power grid voltage instability identification system, comprising:

[0030] Key response bus determination module: Implement voltage partitioning based on reactive power flow, determine the partition section based on the principle of the most drastic voltage change, and thus determine the measurable key response bus;

[0031] Index calculation module: obtaining the voltage timing signal on the key response bus, introducing a correction function to prevent the Lyapunov exponent from crossing zero due to errors, and calculating the improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from crossing zero;

[0032] Scenario identification module: According to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the voltage timing signal on the key response bus, the scenario identification result of voltage instability caused by the control system or load-side voltage instability is obtained.

[0033] The third aspect of the present invention provides a computer device, comprising: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, a response-driven new energy AC and DC power grid voltage instability identification method is executed.

[0034] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, a response-driven new energy AC / DC power grid voltage instability identification method is executed.

[0035] One or more of the above technical solutions have the following beneficial effects:

[0036] The method proposed in the present invention for distinguishing different voltage instability scenarios by using the improved index of anti-Lyapunov index does not rely on the system model, is suitable for identifying instability scenarios in non-plan-based strategies, and can improve practicality. Compared with the offline time domain simulation method, this method has a fast calculation speed. When a complex fault occurs in the new energy AC / DC hybrid power system and the table lookup action cannot be controlled by the plan-based control, this method can quickly identify and distinguish the voltage instability scenarios based on the response.

[0037] The present invention selects a key response bus and takes the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the key response bus voltage timing signal as the key response characteristic quantity, and performs voltage instability judgment and instability scene identification according to the change law of the improved index. The new voltage instability scene caused by the control system in the AC / DC hybrid power system containing new energy can be distinguished from the traditional load-side voltage instability scene, and is no longer oriented to a single scene.

[0038] The present invention focuses on solving the problem of distinguishing new voltage instability scenarios from traditional load-side voltage instability scenarios in new energy AC and DC power grids, as well as the problem of missed judgment of voltage instability scenarios by traditional criteria. It can serve as an important basis for analyzing voltage instability factors and provides a research basis for taking targeted control measures.

[0039] The present invention takes into account the influence of the measurement error of the synchronous phasor measurement device on the discrimination result, and proposes a correction function to improve the accuracy and error resistance of the recognition method.

[0040] Advantages of additional aspects of the present invention will be given in part in the following description, and in part will become obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0042] Figure 1 This is a flow chart of a method for identifying voltage instability scenarios of a response-driven new energy AC / DC power grid in Embodiment 1 of the present invention;

[0043] Figure 2 This is a structural diagram of a new energy AC / DC power system in Embodiment 1 of the present invention;

[0044] Figure 3 is a simulation curve of the bus voltage when the load side voltage is unstable in the first embodiment of the present invention;

[0045] Figure 4 is a simulation curve of bus voltage when voltage instability is caused by a DC control system in the first embodiment of the present invention;

[0046] Figure 5 is a simulation curve of bus voltage when voltage instability is caused by new energy control in the first embodiment of the present invention;

[0047] Figure 6 is a simulation curve of the Lyapunov index improvement index when the load side voltage is unstable in the first embodiment of the present invention;

[0048] Figure 7 This is a simulation curve of the Lyapunov exponent improvement index when voltage instability is caused by a DC control system in the first embodiment of the present invention.

[0049] Figure 8 This is a simulation curve of the Lyapunov index improvement index when voltage instability is caused by the new energy control system in the first embodiment of the present invention. DETAILED DESCRIPTION

[0050] It should be noted that the following detailed descriptions are exemplary and are intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0051] It should be noted that the terms used herein are for describing specific embodiments only and are not intended to be limiting of exemplary embodiments according to the present invention.

[0052] In the absence of conflict, the embodiments of the present invention and the features of the embodiments may be combined with each other.

[0053] Embodiment 1

[0054] like Figure 1 As shown, this embodiment discloses a response-driven new energy AC / DC power grid voltage instability identification method, including:

[0055] Implement voltage partitioning based on reactive power flow, determine the partition cross section based on the principle of the most drastic voltage change, and thus determine the key response bus that can be measured;

[0056] The voltage timing signal on the key response bus is obtained, a correction function is introduced to prevent the Lyapunov exponent from fluctuating at zero crossing due to errors, and an improvement index for preventing the Lyapunov exponent from fluctuating at zero crossing of the voltage timing signal on the key response bus is calculated;

[0057] According to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the voltage timing signal on the key response bus, the scenario identification results of voltage instability caused by the control system or load-side voltage instability are obtained.

[0058] The proposed method of this embodiment aims to extract key response characteristic quantities based on the difference between voltage instability on the load side and voltage instability on the new energy side or high-voltage DC side caused by the control system, and use the Lyapunov index to improve the index to identify the instability scenario.

[0059] This embodiment is based on the identification method of multiple scenarios of voltage instability of new energy AC and DC power grids, and the main implementation steps are as follows:

[0060] Step 1: Identify the key measurable response bus

[0061] The grid structure is mapped into a graph, in which the busbars are mapped as nodes and the lines are mapped as edges. The edges in the graph are weighted by reactive power flow or a function based on reactive power flow. An adjacency matrix is ​​constructed and clustering is performed based on the spectral clustering method to form different partitions with high bus voltage correlation in the same area and low bus voltage correlation in different areas. Since the edges weighted by reactive power flow reflect the degree of change in voltage amplitude on both sides of the line, the lines on both sides of the partition section are the key response buses for measurement.

[0062] Among them, spectral clustering is divided into two steps. The first step is to map the spatial coordinates, and the second step is to implement clustering. In the second step of clustering, based on the association matrix W, the degree matrix D and the Laplace matrix L can be constructed, and the eigenvector f corresponding to the smallest k1 eigenvalues ​​of the Laplace matrix is ​​calculated, where k1 is the expected number of partitions; the eigenvector f is standardized to obtain the characteristic matrix F, and each row in F is taken as a k1-dimensional coordinate, and then clustering is performed to obtain k1 partitions, the cross-sections of each partition, and the busbar numbers contained in each partition.

[0063] Specifically, the reactive power flow transmitted on the line reflects both the correlation degree of the bus voltage at both ends of the line and the severity of the bus voltage change at both ends of the line. The smaller the index is, the higher the correlation degree of the bus voltage at both ends of the line is, and the smaller the voltage change is. Therefore, the bus voltage correlation matrix can be constructed using this index, as shown in the following formula:

[0064]

[0065]

[0066] Where n is the number of system buses; w ij is the correlation coefficient between busbars i and j, Q ij It is the reactive power flow transmitted between busbar i and busbar j. Here, the inverse of the reactive power flow is taken as an example to assign weights to the edges and the association matrix in the graph.

[0067] In order to classify buses with high voltage correlation into the same voltage partition, Q ij How to separate Q when the grid voltage is partitioned when the two busbars are larger ij The two larger buses are a group optimization problem, and the optimization goal is to make the bus voltages in the same partition more correlated, while the bus voltages in different groups are less correlated. At the same time, this optimization goal is also equivalent to the voltage changes on both sides of the partition section being more drastic. From the above analysis, it can be seen that partitioning is an optimization problem based on the reactive power flow between buses, which can be solved by the spectral clustering algorithm.

[0068] The clustering logic of the spectral clustering algorithm is to make the clustering objects in the same class more similar, while the clustering objects in different classes less similar. When the spectral clustering algorithm is used to solve the voltage partition model, Q ij The inverse of defines the relevant elements in the incidence matrix, as shown in the following formula:

[0069]

[0070] L=D -1 / 2 (DW)D -1 / 2 (4)

[0071] According to formula (1) and formula (3), the correlation matrix W can be obtained. The degree matrix D can be constructed through the correlation matrix W. The diagonal elements of the degree matrix D are the sum of the non-diagonal elements of the corresponding rows of the correlation matrix W, and the non-diagonal elements of the degree matrix D are 0. Then, the Laplace matrix L is calculated according to formula (4), and the eigenvector f corresponding to the smallest k1 eigenvalues ​​of the Laplace matrix is ​​calculated, where k1 is the expected number of partitions. The eigenvector f is standardized to obtain the characteristic matrix F. Each row in F is regarded as a k1-dimensional coordinate, and then clustering is performed using a clustering algorithm to obtain k1 partitions, each partition section, and the bus number contained in each partition. The lines on both sides of the partition section are the key response buses for measurement.

[0072] Step 2: Calculate the improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from crossing zero point fluctuation.

[0073] According to the changing characteristics of electrical quantities in different instability scenarios and combined with the traditional power grid transient voltage stability assessment system, the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the bus voltage time series signal is taken as the key response characteristic quantity.

[0074] The Lyapunov index reflects the stability of the trajectory in the state space, that is, the average exponential rate at which nearby trajectories converge or diverge. If the bus voltage change after the disturbance occurs is mapped to the average rate of change that converges or diverges exponentially over time, the positive and negative signs of the Lyapunov index can be used to characterize the convergence or divergence characteristics of the voltage dynamic response trajectory. The convergence of the dynamic response trajectory means that the Lyapunov index is negative, which means that the voltage is stable, otherwise it is unstable.

[0075] By selecting a suitable reference voltage, the collected voltage time series signal is converted into a corresponding voltage amplitude deviation time series signal. A suitable time window is selected, and the anti-Lyapunov exponent zero-crossing fluctuation improvement index corresponding to each moment is calculated according to the voltage amplitude deviation at each sampling moment.

[0076] Specifically, it is assumed that the power system voltage time series data obtained by the synchronous phasor measurement device PMU at the key response bus is V t ∈R n , that is, V t = {v t,1 ,v t,2 ,…,v t,n}, where n is the number of key response buses (assuming that all are equipped with PMUs), t = 0, Δt, 2Δt, …, MΔt, Δt is the number of sampling intervals of voltage timing data, and MΔt is the total sampling time.

[0077] Calculate the voltage amplitude deviation Δv of each key bus at time t based on the voltage timing signal t =v B -vt , v B is the voltage reference value, v t is the voltage amplitude at time t. Considering that the bus voltage amplitude deviation can be mapped to change exponentially over time, the voltage amplitude deviation at time t+1 is shown in the following formula (5):

[0078]

[0079] Assuming that the voltage deviation is expressed in exponential form, if the classical Lyapunov exponent is calculated, it is as follows:

[0080]

[0081] By analogy, the voltage amplitude deviation at time t+n and the corresponding classical Lyapunov exponent are shown in the following formula (7):

[0082]

[0083]

[0084] Taking time kΔt as the starting time and the time window length NΔt, the classical Lyapunov exponent at time kΔt is as shown in the following formula (8). Let k increase continuously, that is, the time window slides, and the classical Lyapunov exponent of the bus voltage at each time is:

[0085]

[0086] Among them, λ(kΔt) is the classical Lyapunov exponent at time kΔt on the critical response bus, and NΔt is the time window length.

[0087] For formula (9), since the measurement device may have measurement errors, even in the voltage instability scenario, the calculated classic Lyapunov exponent may fluctuate around zero due to measurement errors, which is not conducive to the conclusion. Therefore, the following improvement measures are proposed to form an improved index for preventing Lyapunov exponent zero-crossing fluctuations for voltage instability scenario identification, and the correction function shown in the following formula is proposed: Enhance the accuracy of the recognition method.

[0088]

[0089] In the formula, Δv (m-1)Δt is the voltage amplitude deviation of the key bus at time (m-1)Δt, v mΔt is the voltage amplitude on the key response bus at time mΔt, ξ is a very small positive value, A is the error coefficient, v' max It is the maximum voltage amplitude change rate from fault removal to mΔt time period.

[0090] After introducing the correction function, the calculation formula of the improved index for preventing Lyapunov index from crossing zero point fluctuation is as follows:

[0091]

[0092] Where λ1(kΔt) is the improvement index for preventing Lyapunov exponent zero-crossing fluctuation at time kΔt on the critical response bus; NΔt is the time window length, v B is the voltage reference value, v mΔt is the voltage amplitude of the key response bus at time mΔt, is the correction function.

[0093] Step 3: Identify voltage instability scenarios.

[0094] In the new energy AC / DC power system, the new voltage instability scenario caused by the control system is quite different from the voltage instability scenario on the load side of the traditional power grid in terms of mechanism and performance.

[0095] In the traditional load-side voltage instability scenario, the disturbance causes the equivalent load impedance to decrease, the power factor to decrease, and the voltage to decrease. At this time, the active power decreases instead. The reduction in active power will further lead to a decrease in the equivalent load impedance and a further decrease in voltage, which manifests as a positive feedback process in which the voltage continues to decrease.

[0096] In the new energy AC / DC power system, when the disturbance causes the DC inverter side commutation bus voltage to decrease, the DC commutation fails, the DC output active power drops rapidly, and the reactive power demand decreases, causing the commutation bus voltage to rebound and enter the active power recovery stage. The active power transmitted by the DC increases. Due to the control system, the trigger angle is larger at this time, the inverter side power factor is lower, the reactive power demand of the DC system increases, and the commutation bus voltage is lowered. When the voltage is lower than a certain value, the commutation fails again. This cycle repeats, forming a new scenario of voltage instability caused by the control system, which is manifested as the voltage amplitude constantly oscillating within a certain range.

[0097] In addition, in the renewable energy power system, if the disturbance causes the voltage of the renewable energy station to decrease, the renewable energy unit will enter the low voltage ride-through process, the active output will decrease, and the reactive output will increase the supporting voltage. When the voltage rises and exits the low voltage ride-through process, the active power will be restored. It can be seen from formula (12) that after the low voltage ride-through ends, under the power recovery control, the voltage of the common connection point decreases. When the voltage is lower than the low voltage ride-through threshold, the renewable energy station will enter the low voltage ride-through again. This cycle will cause voltage instability, forming a new scenario of voltage instability caused by the control system, which is manifested as the voltage amplitude oscillating continuously within a certain range.

[0098]

[0099] Where U PCC is the voltage at the common connection point, U S is the equivalent voltage of the power grid, P is the active output of the new energy station, Q is the reactive output of the new energy station, and X is the reactance of the line between the public connection point and the grid connection point.

[0100] According to the characteristics of voltage changes in the load-side voltage instability scenario and the voltage instability scenario caused by the control system, the instability is judged and the two instability scenarios are distinguished by selecting a suitable time window and calculating the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation at each sampling time from the voltage time series signal. When the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation is greater than 0 and the key bus voltage and duration meet the threshold conditions shown in formula (13), it is judged as the load-side voltage instability scenario; when the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation repeatedly crosses zero, it is judged as the voltage instability scenario caused by the control system.

[0101]

[0102] Among them, t is the duration of the Lyapunov index zero-crossing fluctuation improvement index curve greater than 0, u is the voltage value of the key bus, T ts is the set time threshold, U ts is the voltage threshold.

[0103] In this embodiment, a regional power grid is taken as an example to verify the effectiveness of the method for identifying different voltage instability scenarios of the new energy AC / DC power system proposed in this embodiment. The regional power grid contains 1 DC system (LCC1), 6 wind turbines (Gen Wind 1-6), 6 photovoltaics (Gen Light 1-6), and an active load of 4516MW, of which the motor load accounts for 65%. The AC system only considers the 500kV voltage level power grid. The system structure is as follows: Figure 2 shown.

[0104] Calculate the reactive power flow between the buses in the system, construct the bus correlation matrix with 1 / Q, use the spectral clustering algorithm to partition the voltage, and use the buses at both ends of the partition section to form the key response bus. Perform time domain simulation in PSD-BPA, set a three-phase ground short circuit fault at line bus1A-7-bus1A-9, and the line is cut off by protection after 0.2s. Through time domain simulation analysis, it is found that the system has load-side voltage instability. The system enters a positive feedback process in which the active power decreases and the equivalent load impedance decreases. The voltage trajectory of the key bus is as follows: Figure 3 shown.

[0105] A three-phase ground short circuit fault is set at bus1A-13 connected to the DC inverter station commutation bus. The fault is eliminated after 0.2s. At this time, the voltage oscillation instability caused by DC control occurs. The DC commutation fails. Due to the rapid decrease in the active power of the DC output and the decrease in reactive power demand, the commutation bus voltage rebounds and enters the active power recovery stage. The active power transmitted by the DC increases. Due to the control system, the trigger angle is larger at this time, the power factor on the inverter side is lower, and the reactive power demand of the DC system increases, which lowers the commutation bus voltage. When the voltage is lower than a certain value, the commutation fails again. In this way, the DC repeatedly enters the commutation failure phenomenon. At this time, the inverter-side commutation bus voltage trajectory is as follows: Figure 4 shown.

[0106] A three-phase ground short circuit fault is set on the line connecting the new energy station Genfeng 3 and bus1A-12. The fault is eliminated after 0.2s. At this time, the voltage instability oscillation phenomenon caused by the control switching of the new energy station occurs. The voltage at the common connection point decreases, and the new energy unit enters the low voltage crossing process. At this time, the active output decreases, and the reactive output increases to support the voltage, causing the voltage at the common connection point to rise and exit the low voltage crossing process. After the low voltage crossing ends, under power recovery control, the output active power increases, causing the voltage at the common connection point to decrease. When the voltage is lower than the low voltage crossing threshold, the new energy station enters the low voltage crossing again. This reciprocating process causes voltage instability. At this time, the voltage trajectory of the common connection point is as follows: Figure 5 shown.

[0107] The voltage time series signals of each key response bus under the above two different instability scenarios are derived, and the improvement index of the voltage time series signal against Lyapunov exponent zero-crossing fluctuation is calculated according to formula (11), where the sampling time interval △t = 0.01s, A = 0.05, ξ = 0.0001, N = 1, and the improvement index curve of the load side voltage instability scenario against Lyapunov exponent zero-crossing fluctuation is obtained as follows: Figure 6 As shown in the figure, the improved index curve for preventing Lyapunov exponent zero-crossing fluctuation in the scenario of voltage instability caused by DC control system is as follows: Figure 7 As shown in the figure, the improved index curve for preventing Lyapunov exponent zero-crossing fluctuation in the scenario of voltage instability caused by new energy control system is as follows: Figure 8 According to the voltage instability scenario identification method, when the duration of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation is greater than 0 for more than 0.3s and the bus voltage is lower than 0.55pu, it is judged as a load-side voltage instability scenario; when the anti-Lyapunov exponent zero-crossing fluctuation improvement index repeatedly crosses zero, it is judged as a voltage instability scenario caused by the control system, so that two different instability scenarios can be distinguished.

[0108] Through simulation analysis, the effectiveness of the response-driven new energy AC / DC power grid voltage instability scenario identification method proposed in the present invention in voltage instability judgment and instability scenario identification is verified.

[0109] Embodiment 2

[0110] The purpose of this embodiment is to provide a response-driven new energy AC / DC power grid voltage instability identification system, including:

[0111] Key response bus determination module: Implement voltage partitioning based on reactive power flow, determine the partition section based on the principle of the most drastic voltage change, and thus determine the measurable key response bus;

[0112] Index calculation module: obtaining the voltage timing signal on the key response bus, introducing a correction function to prevent the Lyapunov exponent from crossing zero due to errors, and calculating the improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from crossing zero;

[0113] Scenario identification module: According to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the voltage timing signal on the key response bus, the scenario identification result of voltage instability caused by the control system or load-side voltage instability is obtained.

[0114] Embodiment 3

[0115] The purpose of this embodiment is to provide a computing device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the steps of the above method are performed.

[0116] Embodiment 4

[0117] The purpose of this embodiment is to provide a computer-readable storage medium.

[0118] A computer-readable storage medium stores a computer program, which executes the steps of the above method when executed by a processor.

[0119] The steps involved in the apparatuses of the above embodiments 2, 3 and 4 correspond to the method embodiment 1, and the specific implementation methods can refer to the relevant description part of embodiment 1. The term "computer-readable storage medium" should be understood as a single medium or multiple media including one or more instruction sets; it should also be understood to include any medium that can store, encode or carry an instruction set for execution by a processor and enable the processor to execute any method in the present invention.

[0120] Those skilled in the art should understand that the modules or steps of the present invention described above can be implemented by a general-purpose computer device, or alternatively, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0121] Although the above describes the specific implementation mode of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without creative work are still within the scope of protection of the present invention.

Claims

1. A response-driven new energy AC / DC power grid voltage instability identification method, characterized in that: include: Implement voltage partitioning based on reactive power flow, determine the partition section based on the principle that the bus voltage change at both ends of the line is the most drastic, and thus determine the key response bus that can be measured; The voltage timing signal on the key response bus is obtained, a correction function is introduced to prevent the Lyapunov exponent from fluctuating at zero crossing due to an error, and an improvement index for preventing the Lyapunov exponent from fluctuating at zero crossing of the voltage timing signal on the key response bus is calculated, specifically: Calculate the voltage amplitude deviation signal on each key response bus according to the acquired voltage timing signal on each key response bus and the determined voltage reference value; Based on the voltage amplitude deviation signal, a correction function for preventing the Lyapunov exponent from fluctuating at zero crossing due to an error is obtained, and an improvement index for preventing the Lyapunov exponent from fluctuating at zero crossing corresponding to each moment is calculated using the correction function; The correction function is: In the formula, is the correction function, ξ is a very small positive value, A is the error coefficient, Δv (m-1)Δt is the voltage amplitude deviation of the key bus at time (m-)Δt, v mΔt is the voltage amplitude on the key response bus at time mΔt; v' max is the maximum voltage amplitude change rate from fault removal to the time period of mΔt, where Δt is the sampling interval number of the voltage time series data; The correction function is used to calculate the improvement index for preventing the zero-crossing fluctuation of the Lyapunov exponent at each moment, specifically: Where λ1(kΔt) is the Lyapunov exponent at time kΔt on the critical response bus; NΔt is the time window length, v B is the voltage reference value, v mΔt is the voltage amplitude of the key response bus at time mΔt, is the correction function; According to the variation law of the improvement index for preventing Lyapunov exponent zero-crossing fluctuation of the key response bus voltage timing signal, the scene identification result of voltage instability caused by the control system or load-side voltage instability is obtained.

2. The response-driven new energy AC / DC power grid voltage instability identification method according to claim 1, characterized in that: Voltage partitioning is implemented based on reactive power flow. The partition section is determined based on the principle that the bus voltage at both ends of the line changes most dramatically, so as to determine the key response bus that can be measured, specifically: The bus voltage correlation matrix is ​​constructed based on the reactive power flow transmitted on the line to characterize the degree of bus voltage change at both ends of the line; The bus voltage correlation matrix is ​​clustered by spectral clustering algorithm to obtain k1 partitions, the cross-section of each partition and the bus number contained in each partition, where k1 represents the expected number of partitions; The lines on both sides of the partition section are key response buses that can be measured.

3. The response-driven new energy AC / DC power grid voltage instability identification method according to claim 2, characterized in that: The bus voltage correlation matrix is ​​determined according to the inverse of the reactive power flow between the two buses, which is: In=[in ij ] n×n , Where n is the number of buses, w ij It is the inverse of the reactive power flow between busbars i and j.

4. The response-driven new energy AC / DC power grid voltage instability identification method according to claim 1, characterized in that: According to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the key response bus voltage timing signal, the scene identification results of voltage instability caused by the control system or load-side voltage instability are obtained. Specifically: if the calculated anti-Lyapunov exponent improvement index is greater than zero and the duration of being greater than zero is greater than the set time threshold, and the key response bus voltage is less than the voltage threshold value, it is judged as a load-side voltage instability scenario; if the calculated anti-Lyapunov exponent improvement index repeatedly crosses the zero point, it is judged as a voltage instability scenario caused by the control system.

5. A response-driven new energy AC / DC grid voltage instability identification system, characterized in that: include: Key response bus determination module: Implement voltage partitioning based on reactive power flow, determine the partition section based on the principle of the most drastic voltage change, and thus determine the measurable key response bus; Index calculation module: obtain the voltage timing signal on the key response bus, introduce a correction function to prevent the Lyapunov exponent from fluctuating at the zero crossing point due to errors, and calculate the improvement index of the voltage timing signal on the key response bus to prevent the Lyapunov exponent from fluctuating at the zero crossing point; specifically: calculate the voltage amplitude deviation signal on each key response bus according to the voltage timing signal on each key response bus obtained and the determined voltage reference value; Based on the voltage amplitude deviation signal, a correction function for preventing the Lyapunov exponent from fluctuating at zero crossing due to an error is obtained, and an improvement index for preventing the Lyapunov exponent from fluctuating at zero crossing corresponding to each moment is calculated using the correction function; The correction function is: In the formula, is the correction function, ξ is a very small positive value, A is the error coefficient, Δv (m-1)Δt is the voltage amplitude deviation of the key bus at time (m-1)Δt, v mΔt is the voltage amplitude on the key response bus at time mΔt; v' max is the maximum voltage amplitude change rate from fault removal to the time period of mΔt, where Δt is the sampling interval number of the voltage time series data; The correction function is used to calculate the improvement index for preventing the zero-crossing fluctuation of the Lyapunov exponent at each moment, specifically: Where λ1(kΔt) is the Lyapunov exponent at time kΔt on the critical response bus; NΔt is the time window length, v B is the voltage reference value, v mΔt is the voltage amplitude of the key response bus at time mΔt, is the correction function; Scenario identification module: According to the change law of the improvement index of the anti-Lyapunov exponent zero-crossing fluctuation of the voltage timing signal on the key response bus, the scenario identification results of voltage instability caused by the control system or load-side voltage instability are obtained.

6. A computer device, characterized in that: include: A processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the computer device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the response-driven new energy AC / DC power grid voltage instability identification method as described in any one of claims 1 to 4 is executed.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the response-driven new energy AC / DC power grid voltage instability identification method according to any one of claims 1 to 4 is executed.

Images